Welcome to the Q&A Forum

Wow, I had never heard of this method of sorting before! Is this something you plan on keeping?

I would not be surprised if this is somewhat related to why Google sometimes finds and ranks some products, only to drop them for others. If the CMS serves it a different canonical version of the website every time it visits, it will index different products at different times.

The subsequent paginated pages are canonicalised back to the first page, but do you have unique product pages still returning 200 OK? I take it you do, due to the wording of your question but just to clarify: To use Amazon as an example, if a page like this is paginated, is a product page it links to still available even if it is listed on a deeper paginated page? If so, Google won't necessarily drop the product page if it doesn't see it linked to by the paginated listing pages, but it might do if it never sees it again due to pagination and canonicalisation. So you might end up with a situation where Googlebot found the Glastonbury coffee mug in January, keeps ranking it well for a few months, has not seen it again by April and drops it. But a white coffee mug set was found a few times in March, so it ranks that instead.

Have a look at this pagination post from one of my former colleagues as well - it has some comprehensive solutions to ecommerce pagination problems.

Hi there,

You may have a more complex pagination issue. My former colleague wrote a good guide to this sort of e-commerce pagination / canonicalisation problem a few months ago. It may qualify as one of the non-boring reads on the subject Do you see a solution here that looks good for you? - http://www.ayima.com/seo-knowledge/conquering-pagination-guide.html

Hi Robert,

The canonical tag is in place on the dynamic URL, meaning that Google is assigning that "page"'s value to the correct, static URL. As such, this shouldn't be a problem for Google. Not sure why Moz is still flagging this, but my guess is that the Moz system is not set up to disregard overly dynamic URLs when they are properly canonicalised elsewhere. The Moz system likely understands that this isn't duplicate content, however, due to the canonical tag. The dynamic URL warning is just being triggered regardless.

If all your dynamic URLs are canonicalised like this to their proper canonical, static version, you should not have a problem with Google.

Cheers,

Jane

Agree with Dennis, if you were relying on this method alone for all your link building, you're going to have a bad time. However, it's perfectly natural for sites owned by the same person or company to link to each other. The best-practice warnings of "nofollow everything!" exist because people abuse and overdo this sort of linking. If you think any of these sites might otherwise be in trouble for any reason, consider being as safe as possible and nofollowing, but I doubt you have to.

Hi Dana,

Just to be clear, what I'm seeing is that if I visit a page like http://www.ccisolutions.com/StoreFront/category/led-stage-lights with 107 products, and I click "next", I do not receive the same URL with different products, but rather the URL changes to http://www.ccisolutions.com/StoreFront/IAFDispatcher with the next set of results. I think I'm just being blind, but how did you block that URL from crawling and indexing? I can't see a meta tag or a line in the site's robots.txt file.

This set-up definitely needs a modern pagination solution - glad Ryan's post at Ayima helped! This current set-up is definitely very detrimental to passing authority throughout the site to all products effectively.

Hi Alexander,

Check out Who Is Hosting This - it's a large community with a lot of reviews, comparisons, etc. Commercially it refers visitors to hosts, but is otherwise impartial (there are links to hosts with good and bad reviews). The "compare" feature might be useful for you: http://i.imgur.com/wyyITPL.png

I hope this helps!

Jane

This really depends on the context and how well the title flows. Would a title like "Real Estate in Santa Monica, LA, Santa Barbara and San Diego" accurately target the phrase [real estate san diego]? Not particularly well but a high quality website with great links could still rank for that query with such a title tag (although it would be far better for said imaginary website to structure its website differently to target each location, but this is just an example).

"Example Location" is probably the better choice, but again - it depends on the readability and relevancy to the page's content.

Hi there,

Glad to help!

Given that I run a commercial site, do you recommend I completely avoid an author bio and only link (where justified) in the article body?

I would avoid linking out from the bio, that's for sure. That was / is a really common way to get a link (or another link) into a guest post and it's crazily easy to spot. If you link, link to a non-commercial property (this is being very cautious, but since guest blogging was Google's favourite thing to penalise last month, extra caution is okay!).

Some site owners have inserted bylines on my articles, eg: "By Jeepster of Jeepster.com" (followed link), making it clear it's written by a guest. Again, is that to be completely avoided?

A lot of folks have struggled with this. For the time being, I might avoid having a followed link out from a byline like that, especially if the page also says "sponsored" or "guest post" anywhere. Links from the post (which is part of the point, of course) should be okay when relevant to the text, but perhaps be cautious of "extra" followed links from both bylines and bios. There isn't much you can do if someone wants to note that the post was written by someone else, but if the post conforms to all the other quality standards of writers simply writing for each other's websites, it should be fine.

There's a reputable site in my industry that openly asks for guest writers (completely unpaid; and it's not one of those built-for-guest-posts websites). Can I safely contribute to this if my articles contain relevant links?

I would say that this sounds fairly safe, as long as the site also ticks the boxes in that list (i.e. you go through and figure out if it's been abused by SEOs and / or by the site owners for money from guest authors, which it sounds like it hasn't). Google has surprised even the tinfoil-hat-brigade with some of the stuff they've penalised or said is not okay in the last year or so, so any links out from sites that offer guest posting options aren't 100% risk-free nowadays, but I guess the idea is to calculate the risk versus reward when deciding. Sorry to not be definitive here!

I've been granted approved guest authorship to a couple of heavyweight sites in my industry, ie, I submit articles to them which are vetted at their end before going live. Is that still safe?

Same sort of thing as the other site really, although this sounds more like the "authoritative figure writing for a different authoritative site" scenario that is common in journalism, blogging, etc. We can't be sure what Google will choose to crack down on next, but they're already getting heat for trying to control how information is shared on the web and trying to crack down on knowledge being shared like this would be a further step too far, in my opinion.

Hi,

If you have a page on a certain procedure, a blog post about that procedure should aim to expand on the existing content rather than repeat it in an attempt to create fresh content. You're naturally going to mention the same topics, but this is normal and Google sees "similar" content all over the web all the time. What you'd be looking to do with a blog to make it both useful for search engines and people, is take a topic where you have basic information on a "product" page and consider exploring the following for a blog post (please excuse a lack of surgery knowledge here!):

1. Gather the latest medical research on the procedure and talk about how it has changed over time
2. Go into more detail about how a procedure is different for men and woman (if it is) and what needs to be considered when treating patients
3. Tips for how to maximise results after the procedure is complete
4. Common recovery issues and how to resolve them or watch out for them

Any new or different information you can include will add to the status of the blog post as a new, relevant piece of content, rather than something that has been repurposed from a product page.

You can link internally in the way described above, but make sure you're always linking with the intention of explaining what the linked-to page is about, rather than with the idea that you are trying to manipulate how search engines view the page.

Given how strict Google is about footer links (with fairly good reason; they have been abused badly so often that footer links with good intentions get swept up in the mix), I would nofollow those links if you wish to keep them because they've brought you qualified traffic.

If the client site is otherwise quite strong, consider looking for another location to link from in a way that seems editorially-chosen and "natural". It's unfortunate that footer links have become so "poisonous", but they truly are best avoided or nofollowed.

This should not cause a problem. As soralsokal says, internal anchor text linking isn't a regular problem for over-optimisation; the only way you'd be in danger of attracting unwanted search engine attention is if you were churning out hundreds of low-quality blog posts, linking internally over and over again to commercial pages with spammy anchor text.

Google is used to seeing navigational links on ecommerce sites, the vast majority of which are "optimised". The optimisation is fine; it's for users. You wouldn't link to your wine page with any other anchor than "wine" because it tells people what the page will be about Google knows this; navigational elements like this are fine as a result.

Hi Tahir,

I would recommend against making this change actually. There is absolutely nothing wrong with the URLs. I can see that you might be considering doing a restructure that makes the type of product more prominent than the location in the subfolder hierarchy, but it isn't advisable to "stuff" keywords into the rest of the URL, e.g. awnings-canopies-brooklyn-NYC.

The existing structure, from my point of view, looks more professional and should not hinder the site's rankings. What is the reason why you're wanting to make the change?

Cheers,

Jane

Hi Ash,

I haven't, but the code for the menu is definitely appearing twice in the source, once under

id="Mmenu" class="show-on-phones SpeedyDropDown"> class="MenuCenter"><a <span="" class="webkit-html-attribute-name">class</a><a <span="" class="webkit-html-attribute-name">="toggleMenu" href="</a>#">Menu <nav <span="" class="webkit-html-attribute-name">class="main-navigation"> nav

id

dnn_MENU2

and once under

<navclass="mega-navigation"><divid="dnn_menu1">I would say that the home page could be a good target for those two keywords, especially given you are prioritising it in key areas like the home page title tag. You have a bit of a keyword cannibalisation issue here right now because you have the internal funeral planning page and the home page essentially targeting the same set of keywords.</divid="dnn_menu1"></navclass="mega-navigation"> </nav>

Hi there,

I would also seek a second opinion - there are many HTML issues that aren't troublesome enough to bother Google, and the claim that "the quality of coding makes it more difficult for Google to index the site and this may be detrimental to ranking" depends heavily on what those coding "issues" actually are. I'd rather see a list of what they say is detrimental than do a full code audit here - I'd be super curious as to what those issues supposedly are.

That list can be used as a starting point to get the second opinion of other development and SEO companies - I would certainly run the list by SEOs as well as coders, as each group knows more about their chosen field than the other. You'll get a good idea about a) what has been done sloppily and b) how Google might treat that sloppiness.

Agreed that going back to the original coders, especially for diagnosis, isn't the best idea, but if you have proof that something they did is sub-par, you might mention it to them. As DJ123 says above, don't expect a high receptive response if you do, of course!

It could be a legitimate point or it could be a way to charge for services - I'd get some full independent quotes. Perhaps even pay someone a small amount for a audit that does not come with it the chance of further work for that person. That is, the auditor has no financial incentive to find something wrong with the site.

I have to disagree that an "SEO specialist" would complete any of these tasks, to be honest. They are all very outdated techniques that are no longer likely to result in any improvement in rankings. You could complete each all of these tasks in one or two days if you went for easy enough resources.

That said, it takes a lot longer and has many times more benefits if you do the following, which could be the modern version:

• Create ten pieces of unique content for specific audiences and promote them on five or six different social networks with the intention of having that content appeal to the user-base of those audiences.
• Contact eleven different bloggers or reporters about the unique pieces of content you've created with the intention of having them feature that content, write about it or otherwise mention / link to it.
• Identify ten different high quality local or niche review or directory sites in the target industry, audit them for quality and relevancy and work out the best place to submit or request a listing.

Content creation on this scale alone could take months. Outreach can take a few weeks with follow-ups, negotiation and discussion. The audit / review process of relevant listing websites might take a week or two given that most directories / review sites are not great in quality and you'd want to find the absolute best resources for your niche and site.

I would look at them for the entire backlink profile, but pay more attention to the followed links when it comes to something like anchor text distribution. Here's why:

If I were Google and I were looking at a car insurance website, and I saw that it had 200 followed links with brand mentions, URLs as anchor text, "click here", etc., then that looks fairly good. However, if the site also has 400 links with "cheap car insurance" as anchor text, does that look somewhat like the site has been doing something like comment spamming? Yup, it does. They make up a telling part of the profile and Google would be silly not to take them into consideration, but they should not directly affect site authority in the way followed links do.

Keep in mind that DA scores are a Moz metric and whilst Moz attempts to replicate how Google operates to the best of its ability, it isn't a Google figure so can't be relied upon to show how Google views a site's authority.

If you've been penalised, I'd most certainly start with removing followed links that are poor in quality. I have not heard of a situation where someone has recovered from a penalty after removing nofollowed links (this doesn't mean it hasn't happened). It's the followed links that Google cracks down on if they dislike them, and the followed links that make a large difference to your rankings.

Property sites use a range of techniques to handle this - I have seen 404s, 410 Gone responses, 302 redirects and 200 OK responses showing a largely blank page (definitely not recommended) whilst browsing a little this morning.

Others leave the listing live but show that it's no longer on the market, e.g. http://www.rightmove.co.uk/property-to-rent/property-29033160.html

It doesn't sound like you can use this last option, although it would allow you to recycle URLs for properties like rentals that often come back on the market.

If you must go with a 404, try to make it useful as Dave says. Can you customise the 404 page, perhaps pulling in information dynamically based upon the listing that was deleted?

Hi Siggi,

There are certainly a lot of instances of "agent" being used on the page which you may be best to cut down on, e.g. using the product name in the roll-over - there are 345 instances of "agent" on that page alone. Don't worry about the Moz grader tool and the apostrophe - Google can understand this, even if Moz has trouble with it.

So for example, on the text over the image for the "Vyztužený saénový korzet Penelope", perhaps use that text only, rather than including "L'Agent by Agent Provocateur" as well, etc.

Hi there,

This is an incredibly broad question that you'd really need to hire a freelance consultant or agency to work on for an extended period for best results The issue of how and where to promote a new website is pretty much a cornerstone of online marketing.

As far as link development goes, I would start with the link building category here on the Moz blog. There is link development and promotion content on there dating back almost ten years (clearly the older stuff might be quite outdated now).

you're are going to need some keyword anchor text from followed links - but not an unnatural amount!

Yeah the ratio of what you need is crazily small right now - I've seen sites with some amazingly good rankings in competitive spaces with nearly blanket brand / miscellaneous anchors, but of course the sources from which they obtain these links are largely relevant to their competitive market / keyword. Like Mailchimp - 10,000 unique linking domains linking with "mailchimp" (plus many more variations below); 28 linking with "email marketing" (all from some questionable badges, mind you) / SERPs right now. The commercial terms will pop up naturally from time to time and that's fine, but to be honest, we'd stopped seeking or being happy about deliberate competitive anchor text a while ago.

Hi Robert,

Disallowing these in robots.txt is the correct thing to do here - this takes away any potential Google problem regarding the login pages. Google MIGHT still index them (but it's not - see here) because blocking via robots.txt says "don't crawl this" rather than "don't index this URL". Subtle difference, but you do sometimes see URLs that are blocked in robots.txt showing up in the index as JUST a URL (no title tag or meta description, since Google was not allowed to crawl the page). An example from your own site is the affiliate sign up page, which is blocked but indexed (see screenshot here).

Moz obeys robots.txt but it's showing these errors, likely because it has seen the links to those redirecting pages somewhere. But Google should not be a problem.

Regarding the HTTPS issue Travis pointed out, check out this search. This shows that you have pages with HTTPS URLs indexed - the top result for me is this page.

You need to implement the canonical tag to point to the preferred version of each URL (HTTP or HTTPS), or implement 301 redirects from one to the other, so URLs like this one are not indexed as they are now.

Cheers,

Jane

Hi Crystal,

Mike is correct  - the Linkscape report will stay the same until there is a new crawl. Sometimes these have been coming in in as little as three or four weeks, and there is lots of effort to make this as fast as possible.

Cheers

These are all great pieces of advice. You might also want to check out Moz's list of recommended companies. A lot of these companies offer link development services and should prove reliable.

Hi Conrad,

Unfortunately every migration is going to yield different results in terms of how well the redirection goes and how long you have to wait to get your rankings back if they suffer. Thomas' experience is fairly typical (and the resources he cites here are good too). It's impossible to say what will happen - a particularly large site (let's say a big e-commerce site for a high-street retailer) might suffer due to the sheer volume of URLs that need to be moved and picked up by Google; a small website may have an easier time. However, metrics like the age and authority of the moving website may well play into how successful the move is as well. As such, it's really hard to say exactly how a migration will go without seeing the sites (and even then, ranking problems can crop up that were unexpected).

Cheers,

Jane

Hi Robert,

Weird, you are right - the canonical tag appears on the majority of pages, but not on a select few, like https://www.bestfitbybrazil.com/Articles.asp?ID=282, which is indexed with its https URL. Is there something different you can see about this URL structure or where it sits in the site that might cause this? Could be worth pointing this one out to the Volusion team too. It appears that no URL with the Articles.asp?ID= structure has a canonical tag.

The other thing I see about this URL by just having a play around is that just changing the query string at the end to 283 as opposed to 282 is bring up the same page: http://www.bestfitbybrazil.com/Articles.asp?ID=283

Unrelated to the initial issue but a lot of the site's content is also duplicated on http://bestfitbybrazil.blogspot.co.uk/, a Google plus page and Tumblr e.g. https://www.google.co.uk/search?q=%22Bia+Brazil+Leggings+LE2854+are+sophisticated+and+versitile+sexy+leggings.+They+fit+like+a+glove%2C+shaping+and+enhancing+your+figure.%22&oq=%22Bia+Brazil+Leggings+LE2854+are+sophisticated+and+versitile+sexy+leggings.+They+fit+like+a+glove%2C+shaping+and+enhancing+your+figure.%22&aqs=chrome..69i57.1069j0j9&sourceid=chrome&es_sm=91&ie=UTF-8#filter=0&q=%22Bia+Brazil+Leggings+LE2854+are+sophisticated+and+versitile+sexy+leggings.+They+fit+like+a+glove%2C+shaping+and+enhancing+your+figure.%22&safe=off

Hi Anthony,

Unless you find a campaign or individual who has been tracking results, either for a specific keyword or for a specific domain, you can't retroactively do this through tools like Moz.

You can use Searchmetrics' domain visibility tool to see how dominant a site has been in its niche over time, however, which gives you a rough understanding of whether it has ranked well or poorly for its keywords over time: http://suite.searchmetrics.com/en/research/domains/visibility

My former agency has an in-house process like this that we used daily (I wrote about how it works in this post, although it has been developed into a tool and is more sophisticated now). It's useful for when you want to look at historic performance but don't have specific ranking reports and Searchmetrics is a great public alternative.

I hope this helps!

Jane

Hi Prashant,

I find that some of these are indexed whilst other are not. An example of a URL listed in the initial site search that won't appear for a site: search of its own is here.

I agree with Michael that your "not found" pages may be part of the problem, even though they are not returning 404s. They should be returning 404s, or at least more information about why the query "activity based classes" near the location returns no results, and what else people can look for / search for within your site instead.

The characters may also be causing a problem with Google not recognising the site: search as being the same as the given URL.

Hi Dana,

I'm inclined to think that this is a problem with how the tools process links to and from PDFs rather than the passing of PageRank through PDFs itself as well, although I am not sure. I would certainly not say no to a link to or from a great PDF if that was the only option, although I have also actively encouraged clients to re-publish their quality content housed in PDF format as HTML and redirect the PDF file to the new source. We're just a lot more sure about how authority passes through HTML files.

Hi,

You shouldn't have to do anything in WMT as there isn't an option for preferred URLs between HTTPS and HTTP (whilst there is an option to say if you prefer http://site.com/ or http://www.site.com/). Google should pick up and acknowledge the 301s if you choose to do this, ranking only your HTTPS URLs.

It used to be the case that this wasn't recommended for SEO, but Google has come a long way with dealing with secure URLs. Just a quick search for inurl:https will show you how many sites are indexed with SSL URLs. Some have done it properly; some have not.

Google handles special characters in URLs, titles, etc. just fine now - a page like http://fr.wikipedia.org/wiki/Château_de_Versailles uses the proper accent on the a, and is displayed properly in modern browsers as well as in Google's results.

There should be no impact on rankings, although it's safe to say that best practice would include using the proper characters, as they certainly benefit usability.

WMT and GA data rarely matches up perfectly, but William's point about the GA code not working for whatever reason is important.

My former agency wrote our own analytics code for one client site. It was a small piece of JavaScript. The client placed it below a very slow-loading piece of conversion rate optimisation JS that took a very long time to load. Most people coming to the website knew what they wanted from the home page quite quickly and we had about 50% of our SEO traffic clicking away from the home page before our code had loaded. As a result, a huge amount of our traffic was missing from our weekly reports: we could only report on those people who'd stayed on the home page for long enough to have our code load, or who had landed on other pages and taken longer to figure out what they wanted or who had come to the page they wanted immediately.

We had our code moved up in the page code. JavaScript processes commands as they appear in the code, so our code loaded "immediately" after it was placed high in the code, and the CRO code did its thing afterwards. Our traffic went back up (after we had been through hours of explaining that rankings and interest had not dropped, so it was absolutely a tracking problem and not a real traffic problem).

Not saying this is happening here, but take a look at the page load times, JS on the page and placement of the GA code, plus raw server logs if you have access to them, and see if there is possibly a tracking problem with GA.

Hi Cole,

Unfortunately there is a solution for this for international duplication but not national. If we were talking about international locations, the solution is the hreflang tag. I'll link to it here just in case it's of use in the future: https://support.google.com/webmasters/answer/189077?hl=en

Nationally, canonicalisation will remove the non-canonical versions from the index and of course from rankings, as Chris has said.

I've looked at medical queries in the past, and Google is very adept at taking IP into account when returning results, the fact that it thinks my IP is located an hour's drive south of here notwithstanding

I would say that re-written content is your best bet if you can't use one page listing multiple locations (highly unlikely) and truly need separate sites for all 13. There can be a little cross-over / duplication without causing too much worry, but I would be concerned that Google is not good enough at a national level to differentiate between duplicates in the same way it can do this for internationalisation.

Hi Robert,

Just to reiterate what was said here - the issue here will be that the home page is much stronger than the internal pages, hence why it's ranking better. This is quite a common problem. I wouldn't worry about the page grader for now, because once you get the right pages ranking, the page grader will show the correct grade for the pages.

Working on the internal linking to those pages, off-page optimisation, those page's content, etc. is certainly going to help. Strengthening the homepage won't hurt either, as long as those internal pages are linked to well from the home page. You can consider linking to them with optimised anchor text, as long as this appears natural and makes sense for a user. For example, if you were working with an insurance site and the home page is ranking for [car insurance] instead of www.site.com/car-insurance/, link to the car insurance page from the home page with the anchor text "car insurance". You'll know instinctively if this is appropriate for the pages / keywords you're dealing with here. You may well have already done this, but thought it best to reiterate just in case!

Cheers,

Jane

Hi Darcy,

I see what you're saying.

My former company did this internally - the process we used was for competitive analysis, but it did provide an estimated traffic score for ALL the relevant keywords in a niche or an entire market. I wrote a little about this process at the beginning of this blog post, so you might find it helpful for a general idea on the subject. Sadly they do not provide this tool for external use: http://www.ayima.com/seo-knowledge/competitive-backlink-analysis.html

So - we were estimating the amount of traffic a site could expect to receive in the pizza sauce market by doing the following things:

1. Using the Google Adwords tool to compile a list of between 50 and a couple of hundred relevant terms in the market
2. Downloading those terms and their estimated monthly search volumes
3. Looking at where our chosen site or sites ranks for all of the terms. We stop crawling rankings after 30 results because no site is likely to get traffic from a lower position than 30th
4. Assigning a probable click-through rate for each ranking the site has. That is, the site should receive around 25% of the predicted monthly traffic for a term if it ranks first for that term. I think we were using 10% for second, 6% for third... we got these figures both from our own data, taken from a lot of client ranking and traffic data, and from leaked search engine click-through figures
5. The end result is a traffic score, which you see in the second column on the chart in the blog post I link to here
6. Done across every term in the market, for every ranking site, we have a list of the strongest performing sites in any one market. The post's example is housing in the UK

HOWEVER: this has proven to be NOT an exact science by any means for true traffic. The use of these reports was more to compare sites' visibility to each other: it's not likely that rightmove.co.uk received 315,000 visits from organic search in the month we ran this report. However, findaproperty.com probably did receive around 80% of rightmove.co.uk's organic traffic based upon their difference in visibility. The actual numbers could well have been lower or higher.

Unfortunately, this is the best method I know of for looking at traffic potential across an entire market. SearchMetrics does a similar market share job as Ayima's internal tool, but again it is for the purpose of site performance review rather than keyword traffic prediction.

Hi Beth,

To echo Andrew, I would be keen to know if your rankings and traffic have also dropped in similar places, rather than just "impressions", which I am assuming you are taking from Webmaster Tools? Impressions being the number of times a site appears, rather than how many clicks it receives. The impressions data from Webmaster Tools is often not regarded as particularly reliable, whilst Google Analytics data is much more reliable and trustworthy.

Are you running ranking reports through Moz Analytics? I'd be keen to know if the drop tied in with any ranking difficulties.

Cheers,

Jane

Hey Flo,

Good news! This went up literally yesterday: http://www.seroundtable.com/google-link-unlimited-18468.html

See the longer discussion here: https://productforums.google.com/forum/#!topic/webmasters/alde4GNOWp0/discussion

This is the first time a Googler has confirmed the lack of a limit on the crawling of internal links

Hi Jason,

If you use the unique ccTLDs and the href lang / rel="alternative" tag, this duplication will be fine. The tag was brought out in late 2011 and tells Google: "just because this content is the same on an Australian site, a British site and an American site, this is okay - it has been done on purpose." You can also use it to point to direct translations, e.g. "this Spanish content is the same as this English content over here, but one is meant for the UK and one for Argentina." Lastly, you can also use this tag as mark up to say "This is French content meant for Canada, and this English content over here is also meant for Canada".

More information about the tag is available here and here.

Cheers,

Jane

Hi Darcy,

If one is using adwords and finds hundreds of related terms, how best to broaden the pages appeal to that large number of other terms?

Generally, Google is quite good at understanding semantics, related terms and meaning, and can deduce (for example) that "zesty pizza sauce" might be a good page for someone who searches for "spicy pizza sauce", even though the two aren't exactly the same thing. For instance, take a search I did this morning: https://www.google.co.uk/search?q=iphone+holder+for+car&oq=iphone+holder+for+car&aqs=chrome..69i57j0l2j69i60j0l2.324j0j9&sourceid=chrome&espv=210&es_sm=91&ie=UTF-8

The first two results have "mount" boldened instead of "holder", because in this instance Google understands that although I searched for "holder", "mount" is the same thing.

To add to the likelihood that Google will pick up a page for relevant terms, you need a decent amount of text content (nothing crazy, but let's say a hundred words or so describing the pizza sauce and its ingredients, including "spices" or similar related terms to the adjectives like "spicy" you'd also like the page to be relevant for).

There is no great return in trying to make one page rank for absolutely everything. Taking another example, SimplySwim's competition page ranks for [racing swimwear] (http://www.simplyswim.com/departments/PopularSearches/CompetitionSwimwear.aspx) but its home page ranks for the similar query of [training swimsuit]. Whether you are looking at just one page for a wide range of queries or multiple pages should be decided on a case by case basis, but a good rule is to decide whether some terms are synonyms or similar enough in meaning / intent to make one resource suitable for a user. In the case of zesty versus spicy, perhaps. Training versus racing swimwear, no.

Just adding to this thought - it really is about context. I suspect that bounce rate is used very carefully and that Google tries to add context, given that a bounce on an informative article and a bounce on an opening hours page are very different. If you believe the pages should either attract further clicks, conversions or longer periods spent on page, then 60% is a worry.

I would never go as far as to say that you will ever be "penalised" for bounce rate - keep in mind that bounce rate should also be easy to fake, either for positive or negative consequences, for a website or a competitor. Google tends to weigh lightly factors that can be easily influenced. It would not be very hard to fake a high bounce rate from real-looking visitors on a competitor's website if you wanted to - you could do it with automated activity or use something like Mechanical Turk. As such, Google can't put too much weight in things that are too easy to manipulate. Hence why they put more weight in backlinks than in on-page mark-up, etc.

Hi,

The drop could be unrelated to your disallowing the account pages (but perhaps check if the CMS allows random query strings, and look into whether it could have created any upon user action, etc. just in case). It's pretty common to see fluctuations in the number of indexed pages, especially with numbers of pages in the thousands or higher. Have you noticed a decrease in traffic from search that you can match with deindexation of pages that were previously bringing in visitors?

Hi Errol,

Yep, these links all come from the same c-block.

As Mary says, it's likely that Google is well aware that these sites are related. I disagree that that would automatically put you at high risk for a penalty, given that interlinking owned / managed websites is common practice, even amongst people who are not doing so for SEO purposes. Unless the network of sites appeared very manipulative, it is likely that the links would be discounted and no benefit passed between them rather than you being slapped with a bad penalty... however, nofollowing the links between each site and your hub site is the best practice when it is done to any extent / large scale, just in case this was viewed as manipulative. Since I do not know how many links / sites we're talking about here, I would certainly say that nofollowing these links would be the safe thing to do.

Even if you were being super-sneaky with the hosting of these sites, disguising Whois information, making changes to your template, etc., Google is incredibly good at figuring out which sites are related to each other. They've been good at this for a long time, so it's always best to assume that they'll know about professional networks like yours and follow best practice with linking.

Hope this helps!

Jane

If the links are nofollowed automatically by the sites' content management systems, they should effectively be ignored by search engines. That said, search engines most likely at least take into account where your nofollowed links come from, and it won't look good to have a large number of nofollowed links from spammy sources. Of course, these aren't spammy sources.

If the links are followed, they pass PageRank, but can also attract low-quality commenters (who don't care if a post is 48 hours or 48 weeks old). You might leave a good, constructive comment as the 4th comment on a post with a followed link through to your relevant site. Comments one, two and three might also be high quality. However, if the links are not automatically nofollowed by the CMS, spammers can and will end up adding endless comments to the thread with their spam links in the future. This isn't great for your site, as it means you're linked to alongside spam and thus part of a bad link neighbourhood of sorts, despite that not being your intention.

Having a selection of comment links should not hurt the site, but do be wary and very selective about where you post these links. Because comment-based link development was such a heavily-spammed tactic (and still is for some people), you need to be very careful about how you link in this manner, even if it is not for SEO purposes.

Hi,

If consolidation is an option, I'd certainly consider it. What I'd be curious about is the indexation and page count of your most successful competitors. I have not worked with a coupon site personally, and I must admit that the 8,000 page number per town does concern me... however, what I'd do is run a ScreamingFrog crawl (http://www.screamingfrog.co.uk/seo-spider/ - you will need to pay for the premium account at $99 to remove the 500 URL limit) for a look at competitors' websites. This will show you the response codes, canonical tags, directives, etc. that others are using. I am not a fan of the idea that if your competitors are doing it, you should do it too, but this will give you a good idea of what is working for sites who manage to rank well for both smaller terms ([jiffy lube coupon post falls]) and big terms ([kohls coupons]).

I would say that 1,000 is preferable to 8,000 if structured properly, but I'd be really keen to know what the rest of your field in vouchers / coupons looks like from an indexed / live URLs perspective.

Hi Tihomir,

Are you planning to give the remaining category a facelift too?

It would be best to include all three categories under the same subdomain (e.g. the "www." subdomain) and place them in folders, e.g. www.domain.dk/category1, www.domain.dk/category2 and www.domain.dk/category3. www2 isn't technically damaging but it's bad from a usability point of view. It's incredibly unlikely to be remembered, for one, and even more likely to be mistyped as www.

Hi there,

The duplication of the products is not highly likely to be causing an issue here given that many ecommerce sites operate like this, but duplicate content was one of the primary issues Panda sought to weed out. It seems as if Panda can be very hard to get rid of, even if you have cleaned up 99% of the issue: you're doing the same or better than similar sites that are not under a penalty but the penalty remains because a certain amount of duplication (or another issue) remains.

Is the problem uniform - i.e. all products that rank well are not duplicated, and all products that have ranking problems are duplicated?

The structure without category URLs is a little abnormal too but what amounts to a flat website shouldn't hold you back completely either.

Hi again,

This is definitely an option - you would probably be best to consider www.name.com/sg/, www.name.com/id/ instead of subfolders however, because subfolders can often be treated like separate sites, just like totally separate domains are. Subfolders tend to inherit much more of the authority of the parent site.

This is something to consider, especially if there is no good reason (like regulations in different countries or local markets with people who particularly like domains targeting them personally) for them to remain separate.

Agree with everyone else here - this is a pretty horrible strategy when done solely for SEO purposes, and usually ends up being very spammy. Most businesses don't have a broad enough catalogue to warrant multiple sites, and the benefit of having keyword-rich domain names (even, or especially exact-match domains like carinsurance.com) is negligible nowadays. Google had to crack down on this, as domain names used to be a very easy to way to rank.

Google is good at figuring out who owns which websites, so unless you are incredibly dedicated with your efforts to hide details, a network of sites like this is likely to be grouped together - Google will probably know they're all yours. Whilst that alone isn't a terrible thing (lots of businesses own more than one domain), Google has seem networks created for SEO purposes like this so many times that the view they'll take of it is dim. Something as simple as the same Local / Places / telephone information would be more than enough to make that connection.

Hi Thomas,

I think your problem is partially duplicate content:

https://www.google.co.uk/search?q=%22Once+you%E2%80%99re+approved+for+a+ReebokONE+profile+and+you+sign+up+for+the+commission+program%2C+visitors+can+purchase+products+displayed+on+your%22&oq=%22Once+you%E2%80%99re+approved+for+a+ReebokONE+profile+and+you+sign+up+for+the+commission+program%2C+visitors+can+purchase+products+displayed+on+your%22&aqs=chrome..69i57.1682j0j7&sourceid=chrome&es_sm=119&ie=UTF-8#filter=0&q=%22Once+you%E2%80%99re+approved+for+a+ReebokONE+profile+and+you+sign+up+for+the+commission+program%2C+visitors+can+purchase+products+displayed+on+your%22&safe=off

This page is virtually identical to a bunch of others in international subfolders, e.g. http://www.reebok.com/sv-SE/reebokonehome/Landing-Page/, http://www.reebok.com/sv-SE/reebokonehome/ (same page but without /Landing-Page/, http://www.reebok.com/nl-nl/reebokonehome/, etc.

It's highly unlikely that Google sees any of these resources as highly valuable on their own, given their duplicated many times. The solution here is pretty simple (in theory) though: the rel="alternative" tag (also referred to as the href lang tag) is meant for the purpose of telling Google that although these pages / subfolders, etc. are duplicates of each other, Version A is meant for the US, Version B for Sweden, Version C for Finland, etc. You can also create, for example, an English and Spanish version of the content for the United States and say: "these two pages are for a US audience but this one is for Spanish queries and this one for English."

Here are some resources about the tag:

https://support.google.com/webmasters/answer/189077?hl=en

http://moz.com/blog/using-the-correct-hreflang-tag-a-new-generator-tool

Essentially, Google may be refusing to pick this page up because it's basically already seen it many, many times.

Cheers,

Jane

Hi Marek,

Very few people get anywhere with the tweet-Matt option sadly

If you received a manual penalty, this has little to do with Penguin updates - the penalty has been handed out by a member of the Webspam team rather than by the algorithm.

What concerns me about your links is firstly how many links point to the site using commercial terms rather than brand terms as anchor text. This is one of the red flags Penguin looks for, but it's also amazingly easy for a person to discover: http://i.imgur.com/INcW11X.png

No backlink profile created "naturally" (and I realise how hard it is to create a natural backlink profile) would look like that. A Googler would take a dim view of that anchor text spread.

Secondly, I'm curious about the sites that link to you using those anchors. I tried visiting them and many of them returned the exact same 500 database error: http://i.imgur.com/lQHEk3p.png + http://i.imgur.com/zpw6YC7.png

All these sites have the same IP address. The other sites hosted on this IP are all down as well: http://www.bing.com/search?q=ip%3A176.67.167.170&go=&qs=n&form=QBLH&filt=all&pq=ip%3A176.67.167.170&sc=0-3&sp=-1&sk=&cvid=fd590e3d130749f290febb6a76973ced

If links were placed on this network of sites all hosted on the same IP, this would absolutely be grounds for a penalty. The weird thing to me is not the penalty but the fact that you were later told you didn't have one.

It's also worth noting that I'd recommend removing those links, penalty warning / loss of rankings or not. There are some other low-quality pages linking to you multiple times with competitive anchors, like http://www.lanaintl.com/all-about-desks-and-its-types. This just looks ridiculously unnatural and manipulative: http://www.lanaintl.com/ - starts off talking about Albuquerque pest control, them keeps linking out to a UK furniture store

You've also got commercial links from sites with identical themes: http://www.house2homefurniture.com/lc-140-xx.html

Link removal is absolutely necessary here, I'm afraid. These bad links all have to go.