Welcome to the Q&A Forum

Hi Den

Your analytics data, so your traffic, bounce rate, dwell time etc. are not ranking factors - Google is quite explicit on this.  The exception being site speed.

Therefore, there's no problem keeping these pages indexed, even if they receive little to no external traffic.  Furthermore, I'd keep them for the long-tail reason - you never know when that long-tail may pick up some converting traffic.  You may as well leave it out there as it's not hurting anyone, especially not your site.

Hope this helps!

Hey Jon

A quick note about analytics - it will always report your homepage as /index.aspx etc.  So, even if you have a mysite.com, www.mysite.com or www.mysite.com/index.aspx structure, GA will report the landing page as /index.aspx.

Therefore, I would choose the URL structure that you would prefer to keep and what you think looks best.  301 the other URL versions to your chosen domain to avoid duplicate content and to pass on all the link equity from those URLs.  If you do redirect the /index.aspx version on your domain a 301 will pass on all the 'strength' so not a lot to worry about there.

When you're done, it's worth checking that all the links on your site link to the homepage in the same way, including your logo if applicable.  Also update your XML sitemap and resubmit to Googe/Bing.

Hope this helps.

Hey there

First question - this is fine. The disavow file stops Google from counting that link as part of your link profile, but it doesn't stop it reporting as linking to your site.  In order for that to happen, you would need to physically break the link.

Second - you're more than welcome to use the domain:spamsite.com command - Google are happy to accept that.  So yes, for a site containing 100 links or more, use the domain: command and you'll be fine.  I've tried and tested this and it's worked for me.

Hope this helps.

Remember it's one URL per line.

If you want to disavow all of geeky.com, all you need to do is:

domain:geeky.com

That's all!

I'd want to optimise the URL to remove those three extra dashes after vietnam as well.

But yes, I would optimise from a user experience point of view - if you see a URL with "-1" at the end of it, you may expect it to be part of a sequence, i.e. "-2", "-3", "-4".  This could lead to some 404 errors that would disrupt the user journey of the website.

I'd definitely want to remove those extra dashes as well, as the URL looks untidy from a user point of view.

Hope this helps.

I could certainly see the value a few years ago.  Now, however, I agree with you.

I can't see it ever being deemed as stuffing, due to it being almost best practice back then.  Now, however, not only is Google much more adept at identifying what a web page is about (and its keywords), the strength a title gives has also been devalued slightly.

In my eyes, this presents a great opportunity to create a compelling title as a call to action.  Include keywords for sure, but use the title to draw people in - mention free shipping if it's relevant, say you're #1 if you're voted so, include your phone number if you offer a local service.

This is how a title can and maybe should be utilised now.  Don't neglect your keywords, but create a compelling call to action.

Hi there Alec

Regarding the canonical tag:

The tag you want to use should be the page that you want to keep, be identified as the original and want Google to rank.  From what I gather, that would be this page: http://www.bereavementstationery.co.uk/funeral-order-of-service

Therefore, on that page, you should have a canonical tag pointing to http://www.bereavementstationery.co.uk/funeral-order-of-service - and on every subsequent duplicate page, you should have the same tag.

The tag effectively instructs Google to say "these other pages are duplicate versions of this URL, so I'm going to ignore those versions, not flag them as a problem, and just promote the original URL.  What's the original URL?  Here it is, it's http://www.bereavementstationery.co.uk/funeral-order-of-service"

Sometimes making a robot appear human actually helps, ha!

An even better solution, which I believe you're trying to implement, would be to 301 redirect all of the URLs containing .html to their equivalent page without the .html extension in the URL.  Not only does this remove the duplicate content problem, but it will also pass the SEO strength of the pages to the new one.

Unfortunately, I'm not qualified to help you there - don't want to recommend the wrong thing and mess up the .htaccess file!  Hopefully someone more proficient in .htaccess redirects will come along and help you out there.

Hope my canonical explanation helps though!

Hi Dan

No, I really don't think this is worth it.

If having the keyword in the latter part of the URL has any effect at all, it is not a very large one.  Moreover, despite 301 redirects passing all of the link juice of a previous page, you're still losing other potential signals, such as referring IPs, anchor text and social signals (if they have an SEO effect or not, having a page with a number of social signals can look pretty positive to a reader).

Does the potential benefit of adding the full keyword to the URL outweigh the potential benefit of what is lost in a redirect?  I think it's pretty hard to say - and I for one don't like making such changes on that kind of information, particularly for pretty trivial ones.

If the on-page content is rich and about swimming clubs, the title and meta tags are present and so on,  I think that's all that you need to succeed for on-page SEO.

Hope this helps.

Hi there

Yes, not only will it affect the pagerank, it will also affect the actual SEO strength/juice/equity going to the page.

For example, I'll just use a random webpage from Reuters: http://www.reuters.com/places/italy - you can see there is a PA of 84.  If I add a random query string to the URL: http://www.reuters.com/places/italy?randomquery - it has a PA of 1.

So that's just adding a query string.  If you change the actual permalink structure itself, you will definitely reset the PA and strength of the page.  To ensure that the strength of the old URL is passed on, you should implement a 301 redirect to your new permalink.

Hope this helps

Hi Moeez

In Google's eyes it will happen within a few days of the new URL and old URL being recrawled.  The DA will update in either the next scheduled database refresh (December 17th) or if not, the one after that.

Oh wow, that's a lot of spam. Very sorry to hear your client was a victim of this.

If the site was hit by an algorithmic penalty (which could be very likely given the amount of low quality spam links the hackers built) that might account for the sudden drop.  That would also explain why your disavow-ing has had no effect thus far.  It can take much longer for the algorithm (the Penguin portion in this case) to update and take into account any disavowed or removed links on that domain, hence why a penalty might still be remaining.

It's pretty much impossible to predict when that might happen - could be tomorrow or another 6 months.  This is probably the biggest frustration many businesses have with Google - the sheer ambiguity on algorithmic penalties.

With all of that in mind, as sad as it is I think you might be better off starting on a fresh domain and taking all necessary precautions to prevent hacking.  I don't like leaving it to chance on when the Penguin algorithm could update, leaving you with an indefinite period of waiting to see rankings come back.  Similarly, we can't even be sure you have an algorithmic penalty, due to Google's vagueness on that matter as well (although I suspect at least one is at play here).

The way I see it, starting on a fresh domain gives you that control back, although I realise that may not be an option.

Hi Ben

That's very much going to be the cause of the problem.  Once that is fixed you should see it appear in the next few days.

Hi there

If you can add "noindex,follow" to those pages, then yes that would be an ideal solution.  I'd double check your pagination code just to make sure it's done properly, but I have seen pages with paginations/different canonicals be indexed before alongside the original in the past (canonical code is a suggestion, not an instruction like a robots command).

One thing I'd recommend doing is do a search for your paginated URLs with the "info:" search parameter, eg: info:http://www.abc.com/jobs-in-delhi-1

If the search result is your original, main page, eg: www.abc.com/jobs-in-delhi - then I see that as a good sign that Google is following your pagination code and so while they might be in the index, they're treated as pagination versions of your main page and so you will avoid anything like duplicate content etc.

But, to be absolutely sure, I think adding "noindex,follow" is a good idea.

Hope this helps.

Hi there

What I would first do is look at two or more product pages with CSS, JS and images disabled, or run them through a "simple search" in the free SEO Browser tool.

That will show you what the page is going to look like to a Googlebot - how similar does it look to you?  Do the pages look 70% identical? 80%?  You could also use the PlagSpotter tool to see just how similar those pages appear.

The point is, it really would be best practice to get as many product descriptions ready before the site goes live - not only to avoid any duplicate/thin content issues but to also help those pages rank outright to begin with.  The Panda algorithm (that deals with thin content) refreshes monthly (perhaps even quicker), so there is a small window there to get all of the work done, if you were to set it live with no product descriptions.  Would you necessarily be penalised straight away?  Or at all?  I honestly can't say as that would be guess work - but I don't like to take chances on stuff that I can't calculate and so I really would recommend you to get as many product descriptions on there as possible.

Have you considered using Mechanical Turk for this? You could outsource the description writing at a very low cost to hundreds of people, dramatically speeding up the process.  If you give clear criteria in your job spec and provide all that a person would need to complete the job, you have a better chance at quality control.  Rand wrote a blog post on this a while back, and a great blog post, albeit on a different topic, on how someone used MTurk and how they set up their job can be found in this CustomerDevLabs post, which is well worth the read.

I hope this helps.

Hey Cole

1. The more you do change, the less risk involved.  Some might tell you that if you change the content enough to pass "copyscape" or other online plagiarism tools, that would protect you from a penalty. I find that to be slightly ridiculous - why would Google judge by those external standards?  The more you can change, the better in my opinion (but I can totally sympathise with the work that entails)

2. Google will know you own the websites if you link them together, share GA code, host them together, contain the same company details and so on - but my question is why would you want to do that?  I think if you tried to tell Google you owned all the sites they would come out you even harder, as they could see it as you being manipulative.

To that point, others will recommend that you only use one domain and target different KWs or locations on different pages/subfolders/subdomains, as it'll look less like a link network. Downside of that is getting Google local listings for each page/location can be a bit of a pain if the pages all come from one domain.

It's not really my place to comment on your strategy and what you should/should not be doing, but suffice to say if you go with individual domains for each location, you should aim to make those domains (and their copy) as unique and independent as possible.

It's also worth remembering that blocking a URL in a robots.txt file does not automatically mean that the URL will be deindexed.  The robots.txt file will prevent robots, such as the Googlebot, from seeing or accessing the file and that's all.  Now obviously, if a robot repeatedly tries to access a file and gets denied it will eventually stop trying to, which is what leads to the file being deindexed (the same principle applies to 404 and 410 errors).

Therefore, if you want a quicker and more definite deindexing solution, you should use the explicit noindex robot command, as recommended above.  This will tell any visiting robot to not index it straight away, which will reduce the number of revisits and drop the page from the index faster.