Welcome to the Q&A Forum

Could be both.

According to SERoundtable, there were lots of reports of a Google update between the 1st and 5th: http://www.seroundtable.com/google-nov13-update-17630.html

Anyway, is a not-confirmed update so it could also be the disavow file doing its work. If some of the links you disavowed were the reason of the ranking drop, then you actually got those rankings with artificially created links. So don't remove them from the disavow file because of the drop, just focus on getting new ones. How? Content!

Hope that helps!

I've done a little more research on this I was intrigued...

My site uses a custom blog, not a public platform, if you visit it (the blog area) it looks alike wordpress but it isn't. We also use NetDNA to serve images, which is the parent company of maxcdn.

My image search in Google returns to my posts, not the images (but there are 2 differences with your site: custom URLs on NetDNA (also possible with maxcdn) and not a Wordpress site).

Then I checked for a small blog I write about SEO, works with wordpress but without NetDNA (CloudFlare instead as the CDN) and the images in google search also link to my posts.

So that leaves with only one option, custom domains. You are serving images using a third party domain, which can be seen as a hotlink by Google and therefore they may link to the domain hosting the image, as it is probably the image owner.

This are just guesses, but it looks like a nice/possible theory.

Why don't you start using custom domains with maxcdn and see what happens?

Wordpress does it yes, but it also offers the ability to link to "nothing" meaning images within the articles won't link to the image source.

That isn't the problem, yet. Your articles' images do not link to their source.

You could use some kind of protection by using .htaccess and a php script. In any case, you should be able to determine the post to which the image belongs.

Something quick (.htaccess):

RewriteCond %{REQUEST_URI} .(jpe?g|png|gif)$
RewriteCond %{HTTP_REFERER} ^http(s)?://(www.)?google.com [NC]
RewriteCond %{HTTP_USER_AGENT} !(Googlebot|Googlebot-Image) [NC]
RewriteRule ^wp-content/uploads/(.*)$ http://www.example.com/php_script_to_find_post_from_image.php?img=$1 [R,L]

Line 1 searches for images; line 2 checks if user comes from google; line 3 checks if it is googlebot; line 4 redirects to a page that ultimately redirects to the post that has the image.

Things to consider:

• This will probably not work having maxcdn enabled.
• This could cause issues with the latest Google penalty "Image mismatch".
• You will need to name each image matching the post, perhaps same title? or you can search all posts that have the image and then decide what to do (as a post can also have more than 1 image and can't be called the same; or an images could be used in more than one post).

Hope that helps a bit.

Most likely using the rel="alternate" hreflang="x" tag, where X is not only the language but also the location. You should have something like this:

1st line tells the default page.

2nd line tells that visitors with french browsers in France should get that page.

3rd line tells that users with french browsers should get that other page.

Think the best way to apply that and you need to do it for each page that has different versions. Those lines are just for the homepage.

PS: That won't guarantee that Google will show the specific page you want, nothing can, but it will help Google understand the structure and serve the best page to the user.

Hope that helps!

Care giving your site's URL?

Within your articles where your images are, do you link to the image source if the user clicks on them?

But all twitter links are shortened to t.co link. Using a custom shortener makes it easier to recognize a "better" source (CNN = cnn.it, Search Engine Land = selnd.com) while allowing you more space for the tweet.

If I would have to choose, I would definitely go with a custom shortener over a free one. Bit.ly even offers custom shorteners for free (if you have the domain).

Oh yes, sorry Luca, I thought you were serving different canonicals in each version pointing to itself. The way you have them set is right.

I don't think this is the reason you are seeing the ranking or traffic drop, as actually using the canonicals the right way (the way you are) should avoid any duplicate content issue. But you better cover all the bases and fix everything that could be the reason.

By the way, did you check the manual actions sections in WMT?

I hope not. Didn't want to say it, but I'm kinda hating analytics :S I wonder if I would remain as a member if the pro version is ever removed.

I feel the same way! Even though I am forced to access Analytics by default, I then hit the "Back to Pro" not only because I am used to it, but also I find it easier to read. I really don't want to spend 10 minutes loading for pages and graphs, and the PRO version gives me just that, the info I need instantly.

As far as I understand, having both versions of the page HTTP and HTTPS can cause duplicate content issues. Serving pages under both protocols creates 2 pages of each page for search engines, even if the content is the same.

Canonicalization here even makes it worth, as you are actually telling engines to index both as different while they are actually the same.

You need to go with one or the other. Given the common choice and your type of business, I would choose HTTPS and 301redirect ALL HTTP pages to their HTTPS. That can be done with a single PHP line of code (if you are under PHP).

My 2 cents.

Just focus on content and a little bit of outreach/PR. Do not make link building. If you have good content, links will come, "build it and they will come".

I started focusing on posting some great stuff on our blog, including the images; a few days ago we got a link for entrepreneur.com because of an image we posted, couple days later 2 links to an article from SEJ and ahrefs. No outreach or PR was done for any of those links.

So, if you ask me, just focus on content, not only the words, images, video, presentations, etc. Links will come. 1 link from a good site, an earned link, is much better than links you may get by doing guest posting for example.

It doesn't really matters. It's just the way you sign in to.

I think the way you have your tags set is perfect. While having a robots=noindex tag allows google to read the page and pass pagerank, there's no point listing them in search results as they actually don't offer any valuable content; they link to, but they don't have it.

That said, google goes over your tag pages passing pagerank, which ultimately get more juice to your articles. Sometimes, within a post, you can link to a tag page within the content, so users can read more about certain topics and you are passing more juice to that tag page, although not being indexed, that tag page passes pagerank back to the articles.

The noindex has nothing (almost) to do with pagerank, it could be a tremendously useful page (you can even earn links to it from external sources), but they are only valuable to users reading your content and interested in a particular "subject" you mention within the post/context.

An easy one here

Moz Analytics:

Go to your campaigns, then click on "Manage Campaigns to the top right of your campaigns table. In the next page you can see a "Delete Campaign" under the column named "Permanently Remove".

Moz Pro:

Go to your campaigns, click on the campaign settings and at the bottom you can find a DELETE button in red.

Hope that helps!

Each TLD has its purpose. If you are an organization, and you are under an ORG TLD excellent, but let's say you offer a consulting service for a fee, then a PRO TLD will probably look better.

In my opinion, nothing works better than a general TLD, com, net, but that ultimately depends on your targeting. Did you try variations of what you need on a .com?

What is exactly what you are going to offer?

Google has released a new penalty called "Image mismatch". Which actually penalizes sites that show images to Google than are not the same as the ones offered to users when accessing the site.

Although I agree with those sites that the image is completely different that the one shown in image search, lately I've seen lots of big sites using some king of watermark or layer that reads something like "To see the high quality of this image, click here" in order to "force" the user to visit the site hosting the image.

Considering the latest changes to Google's image search, which made lots of sites lose their "image search traffic", are these techniques considered part of the new penalty Google is applying? Or does it only apply to the first scenario when the image is completely different?

You can read more on this new penalty here.

How about setting some GET variables to differentiate options in the URL? I saw that on Amazon, when you change options, sometimes the page contents reload with the selected option, showing not only the selection but also some additional info about that particular selection, say images on selected colors, descriptions for that color, etc.

It could be a hard work to complete, but it will ultimately lead to more pages and probably more listings in Google. You should also use the "URL Parameters" under "Crawl" in Google Webmaster Tools to let Google know what those URL parameters do. For example if a get parameter "color=blue" then you can tell Google that the "color" parameter "Specifies".

You can read more here: https://support.google.com/webmasters/answer/1235687?hl=en

Hope that helps!

All the methods you mention are valid as long as you nofollow them.

And if having one domain to redirect simplifies your work, then go for it.

Remember to always nofollow those links.

With every site redesign and restructure comes a little drawback in search engines. 2 weeks is too soon to judge, but the fact that the new URLs are being indexed,, rich snippets are showing, etc is a good sign.

There's no such thing as a penalty for a site restructure, you just need to wait while Google catches up.

Now you should be focusing in conversions/leads. Is the new site appealing to your customers? are you selling more? are you collecting some kind of feedback? That's what really matters, users.

Hope that helps!

2 ways to do it with different effects:

1. Robots.txt in each subdomain. This will entirely block any search engine to even access those pages, so they won't know what they have inside.

   User-Agent:*

   Disallow: /

2. noindex tags in those pages. This method allows crawlers to read the page and maybe index (if you set a "follow") the pages to which you link to.or "nofollow" if you don't want the linked pages to be indexed either.

Hope that helps!