Welcome to the Q&A Forum

Hey Daniel

You might be interested in reading this article and this thread on Inbound.org that brought this up a couple of weeks ago.

Too early to say what the potential ramifications might be, but I would say it's a safe assumption to make that Google is focusing more on more on promoting trustworthy brands.  Bringing the brand name to the front often helps the user navigate through the SERPs, as opposed to the SEO optimised titles that we see out there.

Think we'll have to wait and see on this one, but hopefully those 2 links can help for the time being.

Hi Aggie

This really could be a number of things, but there's a few places to start looking.

First, I would jump into your webmaster tools and see if you have had a spike in errors, which you can find in the traffic tab.  If the Googlebot is seeing 404 or server errors, your users might be getting the same thing.  Look for any spikes in those graphs and see if there is anything that might put a user off.

Next, I would look at segmenting your analytics data.  Are you looking at all of your traffic as a whole? It may be worthwhile to look at the data over time in segments, such as the non-paid search, referral, and paid search segments Google analytics provides.  It may be that there is one particular traffic channel that is driving the "bad" traffic - such as PPC ads that aren't converting or referral traffic from a new partner.  For instance, one of my clients places a banner on an affiliated website with its own tracking URL.  We saw traffic increase over a number of weeks, but bounce rate, dwell time and pages per visit fell and it turned out that the users referred from the banner were, by in large, bouncing straight off.  After running it for a bit longer, we decided to remove the banner as it wasn't driving relevant traffic.  If you see anything unnatural in these segments, you might be able to identify the cause.

Similarly, have you started ranking for any "educational" terms lately?  By that I mean, have you started ranking for terms like "what is [keyword]" or "[keyword] explained"? If so, you may be getting traffic for the terms, but because your landing pages have served their purpose, users are arriving, reading, and leaving.  Not a lot you can do in this case to improve pages per visit, but you could be able to capture some lead data by adding an email signup to those landing pages.

You should also have a look at your user journeys in Google analytics, which you can find if you go to Audience -> Visitors Flow in the left hand side navigation.  See if you can see a pattern here - are there particularly pages that users are going to that has seen a jump in exit rate over time?  Look at those pages - has any HTML/javascript code gone wrong? Is anything broken on those pages? Are they, in fact, just low quality pages that you could improve?  This might also reveal something.

Don't be too worried though if you can't see anything standing out.  Sometimes these things happen and recover - other times user behaviour simply changes. These methods above might be able to point out any faults you need to address or any improvements you could make.  But if your traffic and conversions are steady, pages per visit decreasing isn't such a bad thing.  In fact, if conversions are steady (how ever you measure them), one could argue that users taking a shorter time (and journey) to convert show that they're trusting your brand and website more to make the decision earlier.

Hope this helps Aggie!

If you're just changing the title tags, then no, you do not need to setup a 301 redirect. You can leave that URL as it is.

As for hops, I would say that anything more than 3 hops and the search engine is not likely to follow it any further.  Here's an article on the topic, complete with a Matt Cutts video on redirects, as well as specifically addressing the hop issue (it's at the bottom of the article).

Hi Adam

Have you seen Gcache+ for Firefox? It's an extension that lets you right click a page and bring up the Google cache version of that page.

All it does is add http://webcache.googleusercontent.com/search?q=cache: to the beginning of the page you're on, which is the google cache URL for that page.  You'll get the time and date from there.

Hope that helps.

Hi Thomas

At the risk of being blunt, they're having you on.

Having used the .net system and a lot of CMS', while I know the inifinite loop problem does exist, it is not a "stock" problem with .net or any CMS I've encountered.  If the 301s are returning a loop, it;s likely the dev team's implementation of the CMS and not the .net framework to blame.  From this point of view (which isn't the whole story of course), it's their job to solve it.

In addition, adding a canonical to one URL would help the problem.  If only one page exists, adding a canonical tag to the one page is a strong directive to Google to index only that version of the URL.  That means, eventually, Google would stop indexing the other URL flagging the duplicate content.  So this would solve the issue.

That's presuming that the other URL that's creating duplicate content is a variant of the page, like a query string (www.domain.com/example and www.domain.com/example?query).  If it was a completely different URL (www.domain.com/anotherexample) and it is showing the same content, but not redirecting, then there would be 2 pages in existence.  Again, either a redirect or canonical will help, although you could also get the devs to add <noindex,nofollow>to the meta tag of the page.</noindex,nofollow>

If all else fails, you could use your robots.txt file to block any crawler from reaching the URL.

Here's a handy guide that helps explain all these options.

Hope this helps.

Hi Douglas

It could be a number of things, but here's a few things to consider.

First of all, what do the keyword search results look like?  I know of a client who was ranking #2 for a keyword - but after position 1 there was a massive amount of local search results before position 2, which understandably soaked up a lot of the clicks.  I'd check your keywords for any local results and see if this could be affecting click through rate.

Second, top 5 is nice, but so often top 3 is even better (as it guarantees your website to appear above the fold, even if there are PPC ads).  I've seen clickthrough rate increase considerably from 5 to 3 and even 4 to 3, so this could be the reason.

Third, double check your keywords using the Google keywords tool.  The local search volume levels are of interest here - it is worth noting that they are an estimation from adwords, not organic search.  If the volumes are low, this might also explain a lack of clicks, although if you're going on industry standard keywords this may not be applicable.  The adwords tool has to be taken with a pinch of salt for organic estimations as well.

Finally, how do your title tags and meta descriptions appear?  Do they have compelling call to actions? They should - these are the primary windows that searches will make their first impression on your website.  If you're an ecommerce site, "free shipping" always attracts the attention.  Similarly, are you utilising rich snippets?  Can you use G+ authorship for the pages you're ranking for?

Just some food for thought and I hope this helps.

To follow up on Andy's point about size - you can submit multiple XML sitemaps to webmaster tools.  When I've used multiple ones before for a large ecommerce site, I believed it helped with the indexing.

It's to be expected that some of your blog posts will have a hook to whatever your selling.  Every single post might be overdoing it slightly, but certainly some would be a good idea.  Not just for sales, but also for SEO.

If one of your blog posts becomes very popular, as in it earns external links and social signals, it will become a strong page on your site. If you link your strong pages to your key landing pages, some of the SEO strength will pass through.  Furthermore, your internal links can use a lot more exact match anchor text than external links, so feel free to link to those pages with the keywords you want to rank for.  Again, don't use exact match every time, but you can use it much more often than with external links.

Hope this helps.

Hi Himanshu

I would not use a 301 redirect or block with the robots.txt file as the dynamic query parameter (basically everything after the "?") may not work if you do that. Your CMS may depend on it, or you may be tracking certain things with the query in analytics (for instance, a client I've worked with uses queries to measure display ad performance). For this reason, I wouldn't redirect.

However, implementing a canonical tag on each page should stop Google indexing any dynamic URLs that are created, which will stop any duplicate content issue. For each page on your site, eg: www.sitename.com, www.sitename.com/page1.html, www.sitename.com/page2.html etc. - have a self-referring canonical tag on each one - basically, include a canonical tag that points to the page you are on. For each example above, that would be:

Get in touch with your dev team or whoever created the website to see if they can implement the tags site-wide.  If you're running Wordpress, the Yoast SEO Plugin is a free tool that will do this for you.

Finally, take a look at the SEOMoz guide to canonicals - it's really helpful and will help to explain my points further.

Hope this helps!

Hi Luke

This tool (http://www.tomanthony.co.uk/tools/bulk-http-header-compare/) will be able to give you the response code each URL is sending, allowing you to see if it is set up correctly.  If anything is wrong, however, it won't say why it's returning that response instead of the desired one.

Hope this helps.

It will. It may take a bit of time for it do so, but they should drop relatively quickly. Google is very good at following the canonical directive.

Hey Marton

Try and think of ways of capturing a wider audience to care about your industry.

Could you run a contest - free personal horoscopes for a year, for example - and look to promote it via social networks, bloggers and buzzfeed type news websites?

Can you think of famous horoscopes on the days of big or prophetic events? Like when a horoscope was eerily accurate to something major that happened in the world? (Think 33 Horoscope predictions that will shock you") Maybe a major sporting event? I wonder what Jordan Spieth's was before winning the Masters?

Or how about contacting a Taylor Swift or other celebrity fan site and offer them free weekly horoscopes on the celebrity every week?

These are just ideas from the top of my head - but in reality they are just variations of the principles that you mentioned - trying to earn backlinks from quality websites, increase your social footprint etc.  The added benefit of these approaches is that they - may - get a bit more exposure and so could result in more referral traffic and general good PR for your site.

No doubt, it's a tricky task - but the good thing about horoscopes is that everyone has one, if they choose to read it! So take a look to see if you could try and capture a wider audience by playing to their interests and incorporating that with your website.

Hope it helps.

Hi Eduardo

It's perfectly fine to put in a 301 redirect to a webpage on a different server.  A number of people redirect entire root domains to completely new websites - you won't be penalised just by doing this.

I'm not sure whether your sites are related to each other, but if they are just ensure that all of the content on them is as unique as possible before linking to your shop domain.

Hi Adam

First of all - very sorry to see this.  Hope you can get them back on their feet.

For something so dramatic, for me it's only ever one of three things:

1. A manual penalty (as notified in GWT)

2. Website has been hacked

3. A penguin penalty

I think with points 1 and 2, you might have already ruled this out, but be absolutely sure. Run every check under the sun to see whether the site was compromised, look for odd URLs that have been created, check Joomla/WP updates (and plugin updates to make sure they're not injecting malicious code) - just anything.

So that leaves us with point 3 - and I know you've mentioned you've looked at the backlinks already, and apologies in advance if this advice is like teaching you to suck eggs.  Have you tried every backlink source, not just Moz, but WMT, Majestic and Ahrefs?  Have you used tools like Kerboo and URLProfiler for quick classification of the sites?  Have you checked the anchor text distribution?

The big calling card for me is the brand name drop - and that the majority of the time relates to a link/spam penalty.  If you want, you can PM me the URL and I'll quickly pop it into Majestic and Ahrefs for a bit of further analysis - if anything just for a bit of sense checking.

Hi there

The process you have there sounds absolutely right.  If the other version of the page exists for a legitimate reason, you should use a canonical tag.  You might also want to add to the head tag of your page, which will prevent Google from indexing the page.

For all other circumstances, a 301 is by in large a good route to go down, as it keeps the URLs  clean and the user journey uninterrupted.  If you have thousands of 301s to consider, you may want to be a bit more stringent in your decision making, as you don't want too many slowing down the .htaccess file.

A timely guide to canonical tags has been released on the Google webmaster page, which you can find here.  Have a read through that, but other than that I say your thinking is spot on.

Hope this helps.

Hi Larry

Couple of questions - is that the only URL for the OOP pages, or are there other versions of the page and/or URL that exist?

If there are multiple pages, then that is definitely duplicate content. However, that can quite easily be fixed. If you add this code to the head tag of all those OOP pages, it will prevent Google from indexing the pages (thus not seeing them as duplicate):

That way you can keep the page for the user but not have to worry about duplicate content.  I would do this anyway even if there is only one version of the page, as the page is thin on content as it is.

If you are displaying that image on other URLs that used to have products on them, but have gone OOP, then those multiple URLs and pages would be duplicate.  Again, if you add the above code into the head text, it removes the problem.  You could also 301 redirect the URL of the product page to the OOP page.  For example, if you had a page for a product called: http://www.audiobooksonline.com/examplerecord.html that is now OOP, you could put in a 301 redirect to the http://www.audiobooksonline.com/out-of-publication-audio-books-book-audiobook-audiobooks.html. page and it wouldn't be duplicate.  You can learn more about redirection here.

Hope this helps.

Hey Rafael

Quite a timely question, as Matt Cutts actually answered this today in his video post:

Here's the video

Generally it's OK so long as the sites are related to one another and it is not done excessively. 50 or more would be seen as excessive.

Here's a summary of the post on Search Engine Land

Hope this helps.

Hi there

In reality, you only have two options:

Rewrite the content on each localised page so it can pass as unique.

Or deindex/canonicalise the localised pages.

I'm afraid that's all there is to it.  If the client wants all of the pages to rank in Google, then they have to provide unique content for each page, or they run the risk of a duplicate content penalty.  There's no other way, I'm afraid.

What might be more achievable is to set up a main page for the products that are offered in the regions, with links going to each region's page.  Those regional pages should be noindexed.  It means those pages won't rank in Google, but on the flip side it means you only need to concentrate on ranking one page. If all your efforts head there, the amount of links and social shares it accumulates could make it a powerful page for your keyword.

I think if the links are broken and Google has been made aware of such, ie it has recrawled and cached the page (simply add "cache:" in front of the URL for the last cache copy - if the URL itself is broken, check if it is still indexed in Google), then it would know that the link has been broken and shouldn't count it.

If that's the case, I don't think the disavow would have any benefit, unless of course if the link were to return, which could be a possibility.

If the page is cached and that cached version has got the broken version = no worries.

If the URL is broken and the page is no longer indexed = no worries.

If the URL is broken and still indexed = check to see if any other links point to that URL (including the URLs site navigation and/or sitemap, if applicable.  If not, should deindex soon.  If there are links, I'd disavow.

Just my two pennies, hope it helps!

Hi Steve

I think implementing a canonical tag here is your best course of action, as you have rightly pointed out.

If you have different URL versions of the same page, like you might with the printer version of the pages, implementing the tag will stop Google from indexing any variants of the URL. Only one URL will be indexed and the dynamic URLs will eventually be deindexed.  That should solve that particular duplicate content problem.

If you have two different pages that need to exist, but they have similar content, you will need to add a canonical tag to one and add the same canonical tag to the other.  By that I mean the second, duplicate page needs to have its tag's URL point to the original.  Again, that should prevent it from being indexed.

More information on canonicalisation can be found with the SEOMoz canonical guide.

Hope this helps!

Hi Matthew

You'll probably be interested to see what Matt Cutts has to say about the issue: http://youtu.be/x0-jw_PfwtY

In short, a small number of sites that are related are fine to link to each other and won't flag quality guidelines.  It sounds like your websites fall under this, so you should be fine.

A problem may occur if there is a gratuitous amount of links or websites all passing PageRank.

If you want to be ultra-safe, you can make the links between the sites nofollow - which will you'll out any chance of a penalty due to interlinking.  But I think you'll be fine without this if done sparingly.

Hope this helps.

Hi Jill

Every so often I hear from someone who gets this sort of ambiguous reply.  It's not that helpful, but here's what I can tell you from past experience.

If there was a manual penalty: the Google reviewer has not mentioned that it is still present.  To me this suggests that any manual penalty that you may have briefly had, if any at all, has now been lifted.

Usually, when you get a reconsideration request that explicitly confirms this, it says that you may need to wait a while until your status is restored.

Because this is also omitted, it leads me to suggest that you may have an algorithmic penalty on your site that is present.  We haven't seen a penguin algorithm refresh for a while, so it may suggest that your site may be suffering from duplicate content and/or Panda issues.

If you scroll to near the end of this article http://www.seowizz.net/2012/10/the-disavow-tool-works-real-sites-real-recoveries.html it features your specific reply and also draws the conclusion that there is no manual penalty, but there is an algorithmic one.

I think it may be a Panda issue, you can cross reference the time of your penalty with the Panguin tool for analytics, which is really helpful.  If one of the last Panda refreshes matches up with your dates, it at least gives you a starting point.

Hope this helps

I'd definitely make that point you made in bold.

If you're a paid contributor, it's a matter of does the income outweigh the drawbacks? It's pretty hard to put a tangible figure on that, but there are definite upsides and downsides.  Arguably it adds to Moneywise's branding to be seen on Yahoo, but you can't track that. What you can track are clicks through to the site.

And of course it all depends on what the goal of Yahoo inclusion is.  If it is just a money-spinner and a worthwhile one at that, don't even put the same content on your site.  It's not worth running the risk of duplication penalties and/or link penalties, depending on how Google sees it.

If it is being done to raise brand awareness then (personally) I think it cannibalises your online visibility more than it promotes it - while still presenting SEO problems.

Outside looking in here, but I hope it helps. I'm with you - it's quite a predicament and a delicate situation, so I hope it works out for you.  At the very least, my SEO advice can be seen as impartial and without an agenda, which may be useful to bring to a discussion among people with the company's interests, plus their teams'/

Hey there

Having a self referring canonical tag, as it were, is something I'd recommend for a couple of reasons.

As you have pointed out, canonical tags can help remove the chance of other duplicate pages getting indexed.  However, Google will indexes URLs and one physical page can have a number of different URLs.  For example, if your website has a search function of anything else that might produce a query string (like domain.com/page?query), these URLs can also be indexed and would be seen as duplicate content by Google.

There would not be a physical page to apply the tag to, but if you add a self referring canonical tag, any dynamic URL that is generated by your CMS should not be indexed, helping to limit the chance of any duplicate content penalty.

Furthermore, I believe that having a tag on your page protects you somewhat from scrapers and people stealing your content.  If your page is indexed first with your tag, any syndicated or duplicate versions from 3rd parties in theory should not be able to rank that content.

You look at places like Search Engine Land and they have these self referring canonicals too.

Hope this helps. Enjoy your weekend!

Hi there Itamar

If I am reading this correctly, are the (now removed) URLs appearing on your own site?

If so, in order to tell Google that these pages are well and truly gone, I would serve a 410 response on those pages, rather than a 404.  This is a response code that tells Google the page has gone permanently, and so it will encourage the crawler not to try to revisit the URL in the future.

That means that any external links pointing to that page will become obsolete and shouldn't be counted in your link profile.  That's the theory, anyway.  For that reason, I would hold off on disavowing those links for the time being.  I'd make a note of it, but if everything goes OK with the 410 response, Google probably won't count those external links towards your site anyway.  Just in case, make sure you get all of the external links and save them in an excel sheet, so if you do need to disavow them in the future, you have them to hand.

I've double checked the onssi.com site in the safe browsing tool (see for yourself here) and it looks as though Google thinks the site is safe - ie it doesn't think it's hacked, which is great.  In any case, you may want to run the site through the malware review process just to be absolutely sure.

Hope this helps.

Hey there

As you'll probably be aware, we have just had the latest Penguin algorithm update, aimed at devaluing low quality and manipulative links.

Of the links pointing to your page, I'd say that probably all of them look manipulative and look to have been built for a search engine.

The sites in question look like link farms - a blog network that links out to a number of unrelated health niche websites (and some random ones).  There's very little value on show for a user, each article looks placed to provide an optimised anchor text backlink to a number of websites.

In addition, each domain looks like it has once expired and been bought up someone and turned into a blog network, probably on the premise that the domains were old and carried PR, DA and PA, so would pass decent link equity.  The anchor text pointing to your website all look targeted as well.

Basically, nearly every link to that page looks artificial, placed for a search engine and intended to pass PageRank, without offering anything for a user. It's exactly the sort of link that this Penguin update would try and target, so it's no great surprise to me that you have lost your rankings.

My advice would be try and get the links removed ASAP.  Last resort would be to use the disavow tool.  You may need to wait until the next algorithm refresh to see the effect of removing/disavowing the links take place - the last time Penguin refreshed was 6 months ago prior to yesterday, so we can only hope the new version will refresh monthly.

In the mean time, I would concentrate on earning links by offering quality content, resources or guides to your users.  Contact webmasters and bloggers in your field that could benefit from your expertise, or alternatively promote the content on your own blog and use marketing, PR and social media to attract attention to it.

Hope this helps.

I don't see any reason why this would be seen as black hat.  On the contrary, I see it as an unintuitive UX idea and you should definitely do it.

The only information your withholding (and you're not even cloaking it) is a price that is dependent on a lot of factors.  You're not hiding any content or links, so there's no worry there.  Even if you were hiding content it wouldn't be a problem, unless it was completely irrelevant and there just to rank the page.

Any affect this could have is that if you're deferring elements to load on the page to improve Time To First Byte, then Google may not read them as they crawl and therefore the content it sees on the page may be depleted, affecting your ability to rank the page.  But for something like deferring a price tag, this isn't relevant at all.

I'd say go for it - think it would be a great idea for user experience.

I'm fairly sure that the Googlebot ignores those

title elements, so I can't imagine this having any positive SEO effect.

The worse case scenario would be that if its got a lot of keywords in the title tag, it could be seen as keyword stuffing and we know that Google can penalise sites for over-optimisation on its content.

However, like I said, I believe the bot ignores these elements.  To be safe, just make sure that each tag is relevant and not overdone or stuffed.  Keep it natural to the reader and there will be no problem.

We have to remember that, for all of Google's progress, it is still an algorithm that relies heavily on links and its internal PageRank.

If your site has links from places with a strong PageRank, that is almost "reason enough" for the algorithm to rank your site (in the very most basic terms).

Of course, the greatest advancement Google has made in recent times is teaching its algorithms (and manual quality assessors) to detect manipulative link building.  They're getting faster and better at detecting them.  Rest assured, that site will inevitably be penalised.

In terms of what you can do, look at it objectively: Would you want to beat them at their own game and do this, but have no guarantee that it would work for you and an almost iron guarantee that it will inevitably penalise you? It doesn't make sense to me - so I would keep doing what you're doing and look to earn more links in the way that you have.  You might improve things now, but down the line your site is going to be in altogether much better place.

Keep at it mate - it's frustrating I know, but it'll turn around.

Purely on the ratio - no. It's no problem at all.  I very much doubt your followed vs non-followed ratio is a ranking or negative ranking factor.

However, all of those no-followed links will not be passing link equity and - I'm very much generalising here - a few spammy links are generally nofollowed, such as blog comments.  I'd just quickly review the links that are there and make sure they aren't spammy, but other than that having a larger amount of nofollow links won't be a problem.

Hope that helps.

If it's any consolation, given that you're in the insurance niche, it is one that Google wants to keep a close eye on and keep clean - so chances are action would be quicker than if you were in the dog accessories industry, let's say!

In theory, absolutely. I just think the disavow tool and its effectiveness is a bit of a grey area.

Call me paranoid, but I'd rather not associate any old domain that has had a penalty, lifted or otherwise, with a new domain in any way shape or form.  Like I say though, that could be overkill.

Parasite hosting is when someone places a webpage on a domain that is not theirs and intends to rank it organically.

You might want to do this if you want to ensure that anyone searching for your brand, for example, reaches a page that you own or created.  For example, you might want to rank a Yelp page, Facebook page or others to control your brand results.

Conversely, others may want to rank for keyword terms by creating a page on a website with a high authority (let's say domain authority), as the page they create will inherit some of the trust and strength of the root domain.  Some people rank those pages in a white hat way, others may choose to aggressively link to those pages - as they don't have to worry about their own site being penalised for it.  You can see there where it becomes a gray area for SEO.

The most extreme example would be people in very spammy health niches, eg viagra, compromising a site and creating a new page to sell their products or promote their shop for a quick rankings boost.  This is just outright black hat SEO and illegal, as it involves hacking into a site to create a page.

So you can see there are different degrees of parasite hosting, each with different levels of ethics. Hope this helps.

Any time! It's definitely wise to err on the side of caution. If you ever want to chew the fat on SEO/digital stuff I'd be happy to help.

Hey P

If you're set on taking action, then being aggressive with them is actually a good approach. Both in Cyrus Shephard's great Moz blog post on the disavow tool and also advice from Google itself says if you suspect an entire domain to be spammy, go ahead and disavow all of it.

However, from my own perspective, I would only go through and create a disavow file if I knew for sure that I was suffering from a manual or algorithmic penalty.  I have seen very little benefit in being proactive with that tool (eg rankings are good, you spot bad links in your link profile and disavow them to be safe) and, in fact, I have seen a number of cases when a disavow was submitted "prematurely" - ie, a site was ranking fine and then disavowed some links and saw rankings fall.

If we want to look at it from a slightly skeptical point of view - if you're not suffering from a Google penalty, do you really want to inform Google that you have suspicious links in your profile?

However, that is a matter of preference based on my own experience.  I would certainly take note of the links you think are bad (and perhaps put together a file ready to go, just in case). Keep in mind though that if you visit the page and the link is not there - and especially if you do a google search for cache:http://www.example.com and see that the cached version contains no link - there's a very good chance that the link has already been discounted anyway and so would not be flagged in a manual or algorithmic check.

Hope this helps

Hi there Chris

If I'm understanding you correctly, I also agree with your conclusion. If you look at JS loaded content like on the bottom of this page (click the "read more" at the very bottom), that is all content that Google can see and parse.

You can also put this to the test yourselves. If you go to SEO Browser, insert your URL and press "simple", it will show you how your page looks to the Googlebot.  If you can see that content that you talked about in the result page, you can be sure that Google sees it too. Definitely one of my favourite tools

Hope this helps!

Hi there

The Google Related Searches section and the Google suggested searches (the search terms suggested when you start to type in the toolbar) are both intrinsicly linked and are generated in the same way.

There are two great resources on how this works and where they come from.  The first is a Whiteboard Friday from a not-so-bearded-Rand.  The second is a great article by Rishi Lakhani on how it works and how one would go about "influencing it".  Heavy on theory, but a great read.

I hope these help.

Hi there

It sounds like you've got all the basis covered.  To avoid any form of duplication, it is important to redirect the old version to the new URL as soon as possible, preferably simultaneously when the new blog is published.  That will ensure the two won't be indexed together, plus it makes sense that an updated would be redirected.

There are other methods as well - you could simply remove the old blog post page and return a 404 (which isn't a bad thing), update the canonicals like you say or add a robots instruction to prevent the page from being indexed.  But I think a 301 redirect here would be best, as it would pass on any link equity (no matter how small) the previous page did earn, it's one of the quickest methods to remove old pages from the Google index and for continuity's sake I think it makes sense too.

The fact that you're changing the old post in a substantial way might mean that you could potentially keep both versions on your site, but only you'll be the judge of how unique it is.  I think, to remove any doubt, you should use the 301.

Hope this helps.

Hi Mike

I'm looking at your link profile now and there are quite a few links in there that could be perceived to be unnatural (what betting site doesn't have these though, right?  Used to work for a big brand in-house myself), however, I'm with you in being concerned about these Russian links and anchors.

It looks like your forum has been getting spammed massively.  I can see russian anchors with anchor text like "useful links" and "useful products" - I can even see some Minecraft anchor text in there.  It almost looks as though someone is using your forum to build forum links to rank their spam site.

Have you had trouble recently with spam hitting your forums?  I worry that some people have been ploughing your forum with spam - and thus your site with spammy links - and it may have had a negative effect.

Hi there

I think you're inkling is correct, this would be an old penalty.

The manual action tab is relatively new, coming up to just a year old this month.  Both myself and some agency buddies have seen sites that have had penalties in there when previously the site owner (and sometimes even the agency!) did not think there was a penalty.

If you have Webmaster Tools set up, you should always receive a message whenever manual action has been taken (bar a system error).  However, in the past, for example, you may not have had WMT set up when a penalty has been applied, or you didn't have access to the account then. Therefore, the message would not be there for you to read, due to it not being sent or deleted.  However, the manual action tab will show any penalties that the site has, regardless of when it was applied, when WMT was set up, or if a message was sent or not.

Therefore, if you're seeing a penalty notice in the tab but no message was received, I believe that means it would be an old penalty - as a new one would have also sent a message as well.

Hope that all makes sense.

I'm afraid it will have to be a case of one or the other - either create unique content for the pages or consolidate them into one listing.

You've nailed the reasons already, so I'll offer a bit of advice into making the decision of which option to choose.  I'd look at your analytics data and see how many times those individual product pages have had users entering the website from them, which you could also break down further to see which users have entered via google shopping.

If there's a substantial amount that you believe has led to conversions, then unique content might be the way forward.  If not, then I don't think it losing the other two listings from Google shopping would be much of a sacrifice.

Presuming your product pages sit under a /products/ subfolder, you could use this code in your htaccess file:

RewriteRule ^products/(.*)$ /anotherdirectory/$1 [R=301,NC,L]

Or to redirect all to the homepage:

RewriteRule ^products/(.*)$ /$1 [R=301,NC,L]

Obviously you can change the words "products" and "anotherdirectory" to whatever is applicable to you.

For more info on htaccess redirection you can read the Moz guide here.

Provided that there aren't an absolute ton of them (like thousands/tens of thousands), then I would 301 redirect them.  If there was a massive ammount, too many redirects in your .htaccess can slow your site, but that's only in extreme cases.

Having said that, if there are no links pointing to the page and the URLs aren't in the sitemap, then eventually Google will stop trying to crawl them (as there's nothing pointing to the URL).  I'd just 301 redirect anyway - any little bit of "strength/juice" that they did have could help another page.

Hey there

Not sure if this is the article you read, but Yoast wrote a great article on blocking jquery and CSS and how it might affect your SEO.

The recommendation was not to block them with robots.txt.  This is reiterated by a Google-r at the recent SMX conference has said:

“We recommend making sure Googlebot can access any embedded resource that meaningfully contributes to your site’s visible content or its layout”

I'd recommend that you unlock these.

Hope this helps.

Your mileage may vary, but I've redirected a lot more than 100 before to a root domain and everything has gone OK.  If you could split it up, it might ease your worries - but as you said earlier, there isn't a relevant page to point these URLs to, so the root domain should suffice.

I would definitely consider what Paul has to say below, which is what I was trying to get at in my second point.  There's also a point that if 404s are a perfectly normal occurrence for websites.  Again, provided that there isn't a gregarious amount (which ~100 definitely isn't), it won't be an issue to return a few 404s.  It only would be if it interrupted a user journey, which I can't imagine /tag/ subfolders would.

Either way mate, I think you'll be fine how you pursue.

Hi John

Just to make sure I'm reading this correctly.

You have your homepage.  You can then click to the "about us" page.  And from there, you can click to the "competition" page.

However, instead of it looking like: www.example.com/about-us/competition, it looks like: www.example.com/competition

Is that correct?

If it is, then I don't think there is anything to worry about.  Here's why:

Think of your URL as merely a cosmetic thing.  Your URL structure does not have to reflect how a search engine or a user reaches that page.  The most important thing in terms of site structure and SEO is how a search engine crawler (and user) reaches that page - IE, how many clicks away from your root domain it is.

As a rule of thumb, you should try to keep all of your important pages no more than 3 clicks away from your homepage - and ideally as few as possible.  How the URL is presented when the user/search engine reaches that point is not relevant**

**This isn't entirely the case - there is some correlation with including your keyword in the URL and higher rankings, albeit a very slight correlation.  For example, some sites have links to their main pages from their homepage which then take you to a URL like: http://www.example.com/keyword/product-a - that doesn't mean that the search engine has to reach the page by going through that subfolder (which may not even exist), but the display URL contains the keyword and so it might be better for SEO (although I think if this does have an effect it is very, very slim).  Conversely, if your main page URL was http://www.example.com/keyword, but it takes you four clicks to reach there, it would not be good for SEO as your page is so far away from the homepage.

The main guidelines here are to:

• Keep your main pages as close to the homepage as possible (via internal links)
• Try to include the keyword in your URL if possible (or at least use friendly, readable URLs)

Now, on the "about us" page, the link that goes to the "competition" page - does it go directly to www.example.com/competition, or is it linked to www.example.com/about-us/competition and then quickly redirects?

If it's the second scenario, that would be an unnecessary redirect and might involve some of the link equity being diluted.  I'd get the agency to look at linking to the www.example.com/competition page directly if that is the case (although I doubt it will be).

I hope that helps to explain some things!

Building links to a YouTube video can help the video to rank higher, for sure.  It's the same principles for a normal website.

However, I don't think you'd get a benefit from embedding a video, that has a lot of backlinks, into your page.  Can't imagine this would have any benefit in helping the page rank.  If it did, you'd see a lot of websites embedding Gangam Style etc into their pages just to rank.

In summary: backlinks can help the video itself rank higher, but embedding the video onto the website won't pass on any SEO effect.

Hi Richard

It is something you should address ASAP.  While I believe that Google is a lot better at recognising 'accidental' duplicate content - IE URLs with URL parameters - and distinguishing it from 'deliberate' duplicate content - just outright stealing someone's work or trying to rank several pages for multiple terms - that is only my assumption.  To be completely sure, let's stop any chance of Google penalising these pages.

I think, in this instance, a rel canonical tag should do the trick.  You can read more on the tag here in Moz's guide. Basically, on the page(s) where you're having this problem add a "self-referring" canonical tag.  For example, if the page was http://www.example.com/blue-widgets/, the tag would be:

Make sure that, when you implement this, the pages that are generated with the URL parameters aren't also creating canonical tags like:

They should all have the original canonical tag.

What this will do is tell Google that "If you see any pages with this tag, we're aware that they might be duplicate, but please only count and index the http://www.example.com/blue-widgets/".  It works just like a 301 redirect in that sense.

I think this would be the simplest solution for you to implement.  If you're having problems, there would be a way of blocking access to pages with certain query/URL parameters by using the robots.txt file, but that could get quite messy.

Hope this helps

Can't see a problem with this at all.  In fact, it's fairly common practice.

Many companies buy domain names that feature their brand + a destination.  Some companies even buy the misspellings of the brand. For example, GTMetrix has bought http://www.gtmetrics.com, which 301 redirects straight to http://www.gtmetrix.com.

Provided these are redirects and nothing more, it shouldn't be a problem.  A problem would occur if your client started building manipulative and spammy links to one of these other domains, hoping that the 301 redirect would mask the suspiciousness and passed on the PageRank.  Just like all bad quality link building, this could cause problems.

Buying the domains and just redirecting them, however, shouldn't be a problem as it wouldn't be interpreted as trying to manipulate PageRank or pass any strength.  It's simply being done to protect the brand, which we know Google loves, if anything.  Should your client warrant an individual site for these locations in the future, you can then reverse the redirect and go for it.

The last thing you would want, in my opinion, would be thin content on these sites or the main site that only serves to attract local search terms and drive traffic to the site, without offering any real value to the user.  I think your potential client would be fine with what he's doing.