Welcome to the Q&A Forum

If you are open to manage your own hosting , Digital ocean allows you to locate your server at different locations around world. They provide the raw VPS, you need to install your own server etc.

I recently faced similar situation (but not to the extend as yours). It could be panda or something else.

First segment the traffic.  Try by source/channels if one source is worse than other. Try location. Is one source is worse than other. Try Type of customers. You get to a place where instead of saying you loosing the traffic, you would be able to say this source, this traffic, this location is having the most impact. It is easier to find when there is sharp drop,

Btw I faced a big drop on Sept 29 on multiple sites. Did you see around same time?

Most of the SEO suggestions (great quality content, long form content, engagement rate/time on the page, authority inbound links ) apply to content oriented site.

But what should you do if you are an aggregator or a resource directory?  You aim is to send the user faster to other site they are looking for or provide ranking about the resources. In fact at a very basic level you are competing for search engine traffic because they are doing same things. You may have done a hand crafted, human created resource that is better than what algorithms are showing.  And your site  likely to have lot more outgoing links than content. You know you are better (or getting better) since repeat visitors keep coming back.

So in these days of Search engines, what a resource directory or aggregator site do to rank? Because even directories need first time visitors till they start coming back again.

Moz spam score should be looked as guidance and not a last word. I had some legitimate sites directories shown as spam as directories tend to have lot more outgoing links then the content on the page. And I would have lost , lots of traffic and link juice I would be getting if I consider spam. And some of the sites are high domain authority and really good research resources.

I do not agree anything above score 14 is "definitely" spam. But it is "likely" spam

I think the  second link is showing as duplicate because the page has exact same title. I did face similar issue and I added "Page=n" to the title using the software. This may or may not be possible in the software you have but essentially the links are not showing the same pages and so there title should be different.

Not sure what exactly outcome you want to have. But I think whatever benefit you will get, won't last long.

If the Website B has a higher DA you are likely get a better benefit from a subdomain.  The redirection benefit will flow to website B since it is ultimate destination. But it has higher DA anyway so you are not getting benefit to website A

It depends on the niche, competition, Quality and frequency of content and also the language. I have been able to rank a site as less as 4 weeks. But Again I could not do scalably across languages and niches.

Unique Quality of content and frequency of publishing goes long way

How do you know their backlinks number is correct? If you are getting that from moz  (or some other tools) that number could be wrong and much smaller than the actual number is. Only Google knows the tru backlink number.

I manage a website which shows way less than number of backlinks in Moz than the actual links shown by google. I am suspecting same may be happening here.

From what I understand (and experienced) There are two separate indices one for desktop and one for mobile.

So the changes you make for each version of the site will go to respective indices and query on each platform will be shown accordingly.  Inside Google Webmaster tool, when you ask google bot to "fetch" there are two separate options. There used to be a time when one of my site used to rank higher on Desktop search but lower on mobile search for same keyword.

So  even though some of the signals may be common, there is a good flexibility is their system to rank sites differently if the format signals are way diverge.

There are multiple online tools which provides this service.Try the URL yourself.

Moz seems to be behind in updating  its index or data. I have fixed similar issue and did 301 redirect a month ago buit moz still complains. So you are not alone.

One such a tool I found is https://monitorbacklinks.com/seo-tools/http-header-status-check

In my experience, Google WMT discovers and reports highest number of links.

And if it has not, then even if other tools reporting some extrenal links but google has not noticed yet, then still you are not getting any benefit from them

It depends on which language customers are searching for.

I have faced very similar situation. It is further complicated by the TLD which is  not from same country users coming from. The website in one language and product description in another.

We gradually added product description in both the language on same page (one below other). Even the  website is in language A, the product descriptions are in language B (with 10% of pages have A and B both). But found that majority of the customers are searching in language B (which is english) and still it ranks well and customers get what they want.  When we see search volume for a product in language A is increasing, we add translation in language A for that product on the top of existing description in language B.

At a fundamental level, you are keeping the data somewhere and it is rendered correctly. In a CMS this data is stored in a database completely outside search engine view. So it does not matter if it is in database or in physical directory somehow. So there is no benefit in keeping the structure same physically.

Having said that and my own experience (we manage website with millions of pages) managing this using HTACCESS script is NOT a good idea. You will be limited by what you can do and maintaining will be quite challenging.

I strongly suggest consider moving to a CMS (like drupal) and store all you content inside a database and the CMS script takes care of HTACCESS plus gives other goodies. There are several tool available to get your content from disk into a database.

In my experience links shared on google+ are indexed quite fast (less than 10 days)

Nothing beats A/B testing.  You may have a great idea but ultimately A/B testing will prove if that works for particular site and customer flow.

1. It is not bad at all. But it depends on what you call content. Some content can make wonders and some may not. Reviews or Testimonials would work best. Do not do just for pure SEO but add value to user experience. I would not add just a keyword heavy text.

2. See the Best buy example below. They have content which helps to choose right cellphone and support plans. Also there is list of Top sellers. It is both merchandising and content simultaneously as helps to show user what is more popular.

http://www.bestbuy.com/site/electronics/mobile-cell-phones/abcat0800000.c?id=abcat0800000

3. Keep in mind mobile shoppers. If your content is text heavy can turn off some customer depending on how it is showing up on mobile device.

Here is another example from newegg.com. They have added customer reviews at bottom. which very clever. This can help you decide at the same time adds keywords for SEO.

http://www.newegg.com/Gaming-Desktop/PromotionStore/ID-2125620?name=Gaming-Desktop

If you have two separate domain names, I would continue to keep them separate and not redirecting one to another. The fundamental strategy I would choose is to gradually move company specific content to brandname.com and keep educational, promotional content on keywordname.com so use that as your thought leadership platform with very relevant links to brandname.com.

over a period you will see people searching for your brand directly going to main company website. People looking for your content will first go to keywordname.com, get educated and most likely gopto brandname.com. By creating your own brandname.com is a great idea since it will create your own traffic stream which is a great competitive advantage. A pure keywordbased domain name is vulnerable to  a better SEO execution by competition so can't last forever.

For SEO it does not help at all.

It is far easier (and can get add SEO value) to create your own new post on google plus. Perhaps you can share the original post and add your own summary with a link. So you are helping it spread and at the same time creating  a new post.

From usability perspective if your comment is really useful and relevant to user you will get direct referral traffic. But if it is not relevant you are essentially spamming

1. It depends on how the content on secondary domains organized. If each secondary domain has a content theme, then it would be easier to get separate links for each of them and thus it benefits everyone. It helps user to quickly find/contribute what they are looking for, helps to attract different specific links and pass the value to primary.  If there is no such theme, then all those secondary domain will compete with each other  for mindshare, user contribution and attracting individual links which would make them to achieve high DA,

2. I have similar setup. but instead of separate domains I have multiple subdomains with content categorized by theme. It helps may ways 1) Easier for single sign on. Use logged in one site does not need to log in again on anotyher site. 2) Can attract different types of links 3) easier for segmentation for advertisers.  Not all subdomains can achieve same DA or traffic but it helps overall network by internal linking.

How are you sure those are "hundreds of variants"?

I faced similar question as I am also seeing several variants of same phrase in Moz. Some can be easily distinguished (hyphan, extra spacing etc) but many look exactly same.   Then I found that in GWT, those variants are not reported as variants. So in other words GWT is only reporting "real" unique keywords while Moz is not able to somehow combine/distinguish what is unique. At least the GWT data  (as seen directly from GWT UI) is clean.

How you are getting your data? It is possible you and Moz may be using some API to get this from Google and there is an issue where several duplicate keywords are getting reported as variants and you left to wonder why they are not exact phrases.

Can you confirm your observation, directly going inside Google webmaster tools?

1. I absolutely think having additional facebook likes increases both trust and conversion.  Social proof as a marketing enhancer has been working for mankind long before online social media came into picture. I have seen it is working in steps (jumps). Having X+10, X+20 may not help by y=X+1000 suddenly helps (1000 is used as example and it would be different for different domains). Once you reach Y, y+10, Y+20 does not help till you goto say z=y+5000

2. However buying on fiverr does not help.  They way I would suggest is to use official facebook boost for pages and posts. Try various combination of targetting etc to see what works for you. Now it is very possible , as you explore this your brand getting more visible to target audience and that is why conversion increases. but it does definitely help.

3. The very first 50 fans are most toughest to get for any page. Specially if you can't ask friends or employees to like a page. In that case you would want to try fiverr till get first 50. Because first 50 are anyway not real "fans" but from employees or wellwishers etc who does not add any value to conversion. Once you get the minimum , you can goto step 2 above. Going to step 2) directly with 0 fans, is very costly.

I have noticed a strange bump in traffic from August 14, coming from Granada Spain. I have a language specific website and there is no way I have suddenly thousands of readers in Granada. GA show the service provider is opera software asa and domain is opera-mini.net

I am wondering the location shown in GA as Granda, Spain is the correct one. Perhaps Opera is proxying the traffic through it proxy server located in Granda.

Or this is spammer trying to download content?

Anyone else has seen this?

Good answer.

I am leaning towards english as subdomain. Currently the majority is English speaking but the demography shift is occurring with more non english speaking audience getting online and expected to surpass english speaking.

But would like to explore a little more. In this scenario, are there any SEO wise pro/cons for choosing english as main root domain vs as a subdomain if other things are equal?

I have multiple subdomain based news website. Each subdomain is in the format languge1.newsdomain.com, language2.newsdomain.com etc. These are not mere translation of one site into another but giving content tailored for the language and culture of the audience it serves. In essence they are related but different.

When it comes to "english" I have two options.

Option1. Create english.newsdomain.com using the same format and use the main www.newsdomain.com as a launching pad (navigation site) which allows the user to choose language and goto appropriate site. This gives the advantage of creating a common brand of newsdomain.com for all languages. There won't be any other content on this www.newsdomain.com

Option2. Use www.newsdomain.com for english content and only allow the list of other sites as footer or part of menu. This gives the advantage of having main domain with content (and links and other benefits) but makes it a bit difficult for users who do not speak english, to find their destination.

Any thoughts which is better or worse?

It all depends on relevancy of content and other signals.  But also for a new site, it is just far harder to rank for US and west European countries. There is so much competition for those countries and less in India so it is easier to rank there. You may not need this , but I am thinking it would be far easier to rank for African countries now  but the traffic you may get is not meaningful for you.

I am sure you may have already taken care of this, but GWT has a setting to signal which country you want to target.

1. If your contacts know you enough (you know them personally or they have opt in to your newslettet) you can send them a newsletter type email (with some regular content) and link to yoour facebook to join. This will work if they see your name and recall "yes, I know this enitity"

2. If you are not that close to your contacts, use your email list as remarketing list and advertise on facebook to show your advt (which will ask them to like your facebook page). Even though it is paid, it can be very effective. Not many people realize that their email list has lot of throw away email addresses, email to which no one reads. With this solution atleast you are paying for ads only when your audience engages.

1. I would first directly go to deep pages and see from the source code, the GA code is indeed called.  Another way to check this (and many other erros) is to install google analytics debug plugin from chrome and visit that page enabling the plugin. The plugin shows on the javascript console if it is sending data to GA correctly and some other script is not exploding before GA data is sent.

2. How do you know for sure the other pages got visit? One way to check is visit those yourself and then see in GA if those visits show up. Those are in behavior->content

Hope this helps.

Many thanks, Timbronz The solution you suggested worked perfectly. The "\x" is a particularly very different insight as just escaping "%" with "" does not work.

Dennis Seymor,

it is finding that way to redirect was the toughest problem as just escaping does not work and this has stumped many.

I have some old valuable followed links from high ranking domains and I noticed from moz reports they are reporting 404.Visually they looked fne but when I clicked on those they indeed were generating 404.

When I researched further they are defined as

My domain.com   Notice there is extra space between "/" and the closing quote.  It turns out it is sending "www.mydomain.com/ " to browsers.

Any ideas How to solve this? If I should put a perm redirect in apache, how do I deal with these "%C2%A0" characters. It seems the issue is happening at more than one remote domain.

1. Create original  content.  Have some unique perspective. There is no substitute. Original is important than "amazing" as what is "amazing" is different to different audience. And if more people are sharing your content who cares if that was amazing to you or your company.
2. Timing: One of single most overlooked factor.  If you try creating/seeding content different time, different days you will see which days/time work for you. To run a test we shared same content, same network but at different times., The difference can be huge. We have seen swings of 1200%  based on time.
3. It gets better as you get some traction. social proof networking effects clearly helps as you get more shares.
4. Don't try to create content for the planet. Identify a niche audience initially and target them