Welcome to the Q&A Forum

Hi all!

For a healthcare website we have setup AMP. Google Search Console suggests to use article markup for several pages and I am not sure if this is correct.

There are two kind of pages:
1. News pages
2. Information pages, for example: symptoms alcohol addiction  or Binge Eating Disorder

There's no doubt the article markup will be correct for the news pages but I am not sure about the information pages. Do you guys suggest to implement article markup on these pages as well or only use this for real news/blog posts?

Hope you can help me out.

Thank you in advance and happy holidays!

Regards,

Anouk van de Velde

Thank you Anthony

Hi Mozzers,

We have a website that has both http as well as https indexed. I proposed the solution of implementing a canonical link tag on all pages (including the login/secure ones).

Any disadvantages I could expect?

Thanks!

Hi Fabrizo,

How long would it take for a PHP programmer to write this code approximately? Since we would have to outsource this I would like an indication to oversee costs involved.

Thanks!

Another way we found helpful is:

1. adding the meta robots=noindex tag to these pages

2. submitting those URLs + sitemap.xml in Google Webmaster Tools

Hope this helps

From now on we want to store all our sitemap.xml over the next years. Its a nice archive to have that allows us to analyse how many pages we have on our website and which ones were removed/redirected.

Any suggestions?

Thanks

Thanks again. I hope Google will come out with some real guidelines on this subject. It  saves us time arguing with third parties.

For now I will get the canonicals fixed.

Allright, maybe a good question for the Google Webmaster Help Forum right?

Hi Andy,

I agree, it does not seem like a logical solution. Do you know of any documentation on this, maybe even from Google? I would like to give some guidelines to my web developer based on a source.

My web developer has pointed the "good" URLs that I use in my internal link structure (top-nav/footer) to another duplicate version of my pages. Now the URLs that receive all the canonical link value are not the ones I use on my website. is this a problem and why???

In theory the implementation is good because both have equal content. But does it harm my link equity if it directs to a URL which is not included in my internal link architecture.

Hi Igrizio,

I'm not entirely sure but could it be that you're redirecting to //index.php instead of /index.php?

You can use this website to see the the statuscode and any redirect locations of a website.

Best of luck,

Sven Witteveen
Expand Online

Hi Tim,

These are the first things that came to mind:

• Find blogs about a specific topic, write down the PA and email address of owner.

• Find link directories relevant to your website, write down PA and email address of owner.

• Get keyword difficulty data for a list of 1000 keywords.

• Crawl website for 404's and give cms access to fix broken links

Goodluck and let us know how it works out!

Sven Witteveen

Expand Online

Hi,

I'm not sure I understand correctly but i'f you write a post about red wine and you put a link in it as follows:

red wine

I don't see anyting wrong with that. You do need to make sure that google will be able to crawl these search pages, otherwise it will be useless. (that basically means that the page should be generated server-side, so no javascript and ajax and such).

In an ideal situation you might want to consider creating a few static pages for your top keywords if you want to rank for them. In my experience a nice clean URL is quite an important ranking factor.

i.e. yoursite.com/top-searches/red-wine.html instead of  yoursite.com/search.html?query=red+wine

Btw, beware of duplicate content if you are going to index search result pages! e.g. resultsfor the queryies redwine,red+wine,wine, and white+wine might be very similar, but have different url's!

Good luck!

Sven Witteveen

Expand Online

Hi Joel,

I'd say a 301 redirect using your HTaccess file is best. However, if you cannot access the htaccess file i'd go with the rel canonical. Otherwise you would be using meta refresh or javascript and the like, which are generally not appreciated by Google. Besides, if the page is truly a duplicate the canonical link tag usually does the trick anyway!

Good luck,

Sven Witteveen
Expand Online

Hi!

Google will only spend 'so much time' on any new domain. The more traffic and links and page authority you get, the more time Google will dedicate to crawling your website. You should also make sure that the site is not slow, as this will reduce the crawling speed even more! See Google page speed for tips on speeding up the load time of your site

Good Luck,

Sven Witteveen
Expand Online

Hi Tim,

Do you have Google Analytics installed? If so it should be possible to find out what's happening! This is what I'd do:

1. Get the 100 best performing keywords for january - march.

2. Get the 100 best performing keywords for the past two weeks.

3. see which keywords you used to get traffic on but currently don't get traffic on.

4. if you still rank for these keywords, it's probably because of search volumes and if you don't rank first page for these keywords it's because of your rankings. (probably).

Obviously it's a bit more tricky if your long tail rankings only dropped slightly. If that's the case you could try to find out for which keywords you used to get >10% of the search volume in visitors. It's higly likely that you used to rank top 3 for these words. If you rank 6th or 7th for these keywords now you probably lost some ground there.

Also you can checkout the search engine optimization feature in Google Analytics in case you have GA linked to your webmastertools.

This is why it's important to document your rankings over time (on lot's of keywords). You could use a ranktracker for that.

Best of luck!

Sven Witteveen

Expand Online

Hi Justin,

I think you raise a very good point. It is known that nofollow links 'leak' linkjuice, because the linkjuice is not passed through the link, but it isn't distributed to other links on the same page either. The juice that would have been passed through a follow link simply vanishes when you add a nofollow tag.

As you said it is also suggested that Google does not count multiple instances of one link on one page. So the question that remains is: Are the second, third etc. instances of links treated as nofollow links (leaking linkjuice), or are they simply ignored (not leaking linkjuice)?

If they are treated as nofollow links and leak linkjuice anyway, you might as well add a nofollow-tag and make sure you don't get penalized for them either. On the other hand, if they are normally simply ignored by Google, but start leaking linkjuice with a nofollow tag, you might be doing some serious damage to your site.

Quite frankly, I don't know which is the case. However, my gut feeling says that the pagerank sculpting days are over so the above reasoning might not be the way to think about this.

I would simply try not to 'overdo' anything to much. Don't have pages with 200+ links all linking to perfectly optimised pages with perfectly optimised anchors. I suspect that internal linking is not the driving factor behind Penguin penalties anyway, but backlinks are.

Looking forward to see what other people think!

Greets,

Sven Witteveen
Expand Online

Hi Jaishree

It's hard to say without looking at the actual website and SERP's but I'd like to share some thoughts:

I would never use a blog to rank for your most important keywords. It will probably not have as high a conversion rate as your product pages, and it's pages are usually not optimized for a single query. Use your blog to rank for the longtail keywords that do not require a single optimized landingpage.

Also social media is not the way to go when you're looking for better rankings (I think). Don't get me wrong, it's great for driving traffic and building your brand but I personally don't think it will do any good for your rankings in the short term.

So what would I do? When on page optimization is finished, I normally have a look at the SERP's and see if I can understand why I rank where I rank. That is, are pages above you technically better? Is their off-site SEO better than yours? Try to analyse this and the next logical step in your SEO strategy will present itself. the comparison feature in OSE is great for this.

One thing that could make a difference is adding a few (only a few!) deep links to your product pages with optimized anchor texts. However, make sure you only do this if you have a strong healthy link profile because otherwise you're bound to be eaten alive by the Penguin!

Best of Luck,

Sven Witteveen
Expand Online

Hi Kinsana,

It's almost always a good idea to start a blog! The more (quality) content you have, the better you will rank. However, I would reconsider your idea for two reasons:

1. First of all, you generally want to concentrate the value of your entire website on one domain. This way your current website will all benefit from the SEO value you build with your blog. I'd therefore put the blog in a subfolder (www.example.com/blog/).
2. I image it could be rather confusing for visitors to have two seperate websites for one company with exactly the same url (apart from the .net/.com obviously). Imagine a visitor landed on your blog via google and returns to your website the next day by typing your adress in the browser. The autocomplete function will send him to the .net version (based on history) and the visitor might not understand why your website seems to be nothing but a blog. If you are going to use two domains, at least make sure it's clear that one is a blog and the other one isn't.

Personally I agree with Adam, just redirect the .net to your .com and place your Blog elsewhere.

Best of luck!

Sven Witteveen
Expand Online

Hi Konstantin,

A rel=canonical in the head section of a html page will not redirect users to another site. The canonical link tag tells google (or other search engines) where the 'original' version of a page can be found and what the preferrred url is. This canonical version is most likely to show up in search engines (although search engines do not always follow the rel=canonical).

If you want users to end up at site2.com/page1 when they type site1.com/page1 in their browser, you need to use a 301 redirect.

Kind regards,

Sven Witteveen
Expand Online

Hi Ray,

This doesn't sound good..

I'm not that familiar with sitemap.xml protocols but have you checked in webmaster tools that understands your sitemap correctly? Also there might be some info in crawl dates and crawl errors in your webmaster tools.

I can't imagine you've been banned from google because of just one link.

Greets,

Sven Witteveen

Expand Online

Yes it can. Whether or not this will in fact hurt your SEO position depends on the backlinks these websites are receiving.

All the links that point to your domains pass linkjuice. This linkjuice is sent to you primary domain because of the redirects. This means that you will lose all backlinks currently pointing at these domains. My advice is to try and figure out how many links we are talking about, and how much link juice they pass to your current site. You could use opensiteexplorer.org or your webmaster tools for that.

However, If you really don´t mind the cost of keeping the domains at all, i´d say just keep it like it is.

Good luck!

Sven Witteveen
Expand Online

Hi Bob,

If I understand correctly you have some pages on your website that generate some report in some other place and immediately afterwards redirect themselves to this new report.

I think it depends on how important these pages are whether or not this is a problem. If for example a landing page returns a 302 status, that´s a problem. However, if the page generating the report and redirecting itself does not need to be indexed it´s not really an isseu.

You might want to add a rel="nofollow" to the buttons and links that point to the page generating the text reports.

I was looking through some old questions of ours and saw that this one is still open. However, seogadget has obviously solved this problem when they introduced their Google Adwords Excel Plug-in!

Indeed!

In case of problematic keyword selfcannibalization you can choose from two options:

1. Combine two landing pages into one of the two, and deleting (or noindex) the other one.
2. Avoid overlap as much as possible by making sure the exact query you want to rank for on page A does not appear on page B (and vice versa).

I think it depends on the situation whether you want to go for option 1 or 2. In your case it seems pretty much okay, but to be sure you could cut down on the word ciclismo on the general shimano page.

Best of luck!

Sven Witteveen

Expand Online

Hi,

I would definately go for merging both of them together. Allthough WMT recently  significantly increased the number of backlinks shown, any extra information you can gather is great. I guess together they provide a pretty good overview of your backlink profile.

Btw, it's ofcourse very interesting how many of the links reported in WMT and OSE overlap, and how many are unique for each one. So if you're going for option 3, please share some of the results with us!

Greets,

Sven Witteveen
Expand Online

I can reproduce this by:

1. Selecting both the custom segment with regex filters and the standard 'all visits segment'.

2. Turning off the all visits segment.

Now I end up with all words being empty like your screenshot. However, when I refresh the page they are gone.

I'd say just click around in your reports for a while and see if the problem goes away automatically.

Good luck!

Sven Witteveen
Expand Online

Hi,

If you connect your analytics to your webmaster tools, you can indeed see some (rough) estimates on Search Engine CTR (and more). This you will find under 'traffic sources'=>search engine optimization.

For comparing current traffic with earlier time periods you can click the checkbox 'compare to previous period' in the date range selection menu. The report will then also show % changes in traffic etc.

On the point of what keywords to use, look around the SEOmoz website for 'keyword research' topics. That topic is far too broad to be answered in this thread. A good starting point might be the keyword research section of the SEO Guide.

Best of luck!

Historical data can indeed not be copied from profile to another. It is therefore a good idea to create such profile's as soon as you start your SEO efforts. When this is not an option for you I'd suggest you make an advanced Segment.

In this advanced segment you can filter the branded keywords and non organic traffic. However, you cannot filter IP adresses in advanced segments. (see the analytics university I suggested earlier for help

Whether or not you can find additional software that will help you monitoring your SEO efforts, I would strongly suggest you have your Google Analytics set up properly. It is often the source of many usefull insight's regarding your website and your SEO campaing. Trust me when I say it's worth mastering this software.

Hi Shashank,

These websites explain how to make profiles and apply filters to it:

http://www.canadianschoolguide.com/setup-website-profiles.php http://groups.google.com/a/googleproductforums.com/forum/#!topic/analytics/uC6l1DLSLgM

It doesn't show how to filter out certain IP adresses so I've attached a screenshot that illustrates that.

You're right that the keywords you typically want to exclude are branded keywords. However you may also already rank 1st or 2nd for some important keywords that are not targeted in your SEO efforts.

For any other Google Analytics questions you can check out http://www.google.com/analytics/iq.html. You can view short easy to understand video's on any topic related to Google Analytics.

Good Luck!

SEO_Profile.png

I think you can come a long way with Google Analytics and a basic rank checker.

1. The basis is to create a filtered profile for Google Analytics which includes only Organic Search Engine traffic and excludes any keywords for which you already rank very well at the start of your SEO campaign (and preferably also exclude traffic from your IP).
2. Look for increases/decreases in traffic and see if you can mach them to A) changes in rankings for the keywords visitors used or B) changes made by you to that aim to affect the click through rate of the pages the new visitors landed on).
3. See how much conversions (subscriptions or sales or time on site or whatever) are made by the visitors that are (probably) attributable to SEO.

It will never be a fully exact science but I personally think that you can lay a pretty good claim to SEO benefits if they are calculated this way.

Sven Witteveen
Expand Online

I did a quick check on both pages:

http://www.deporvillage.com/ciclismo/zapatillas-shimano
and
http://www.deporvillage.com/shimano

Both get an A for on-page optimization for the word: shimano. Also on the page /shimano there are 41 instances of the word ciclismo. You might want to avoid this word on the general /shimano page if you have a separate landing page for 'zapatillas ciclismo shimano'.

However, generally speaking there isn't much to worry about if you do not in fact rank with both pages for one keyword.

Sven Witteveen
Expand Online

Rand has said the following about it in his blog:

....Keyword self-cannibalization - the practice of heavily targeting the same keyword phrase on multiple pages of a site....

Having multiple pages for one keyword or phrase is not ideal since this means you are competing with yourself. All the backlinks, content, internal linkjuice etc. is devided over multiple pages instead of concentrating it on one page.

Ofcourse, if these pages are on position one and two in the SERP this situation is perfectly fine. However, if they hold position 11 and 15 the combined force of these two pages may be enough to rank 5th or 6th. That would generate a lot more traffic since results after postion 10 generate virtually no traffic.

I hope this helps!

Greets,

Sven Witteveen
Expand Online

Hi Jack,

All though I do agree with you in general, I think this case is slightly different:

1. According to opensiteexplorer.org there are no backlinks to this page containing the word mussoorie.
2. The linkprofile of this page looks very unnatural (almost no branded/url anchors and all are very nice search queries). It is thus very likely that they purchased these links (i.e.use black hat SEO).

Greets,

Sven Witteveen
Expand Online

Hi wildhawk,

All websites on the first page for this query have quite a high PA and DA. They have thousands of backlinks (probably payed for them considering the anchor texts) and are also ´quite relevant´ for the search query.

The basis is this:

You rank based on your on-page optimization (relevance) and off-page optimization (importance). You may have a page that is far more optimized for a specific keyword and thus are also more relevant. That doesn't mean you outperform the 1-10 results on the combined result of relevance + importance.

Long story short: My guess is that you need to work on your linkprofile.

I wish you all the best and keep in mind that you are trying to compete in an extremely competitive market.

Sven Witteveen
Expand Online

It is hard to say what you did wrong without actually seeing the segment filter. However, you might want to have a look at the type of match you used. E.g. If you use 'contains'  'my query' it will also match 'my query with longtail'. Also, offcourse you need to make sure the segment is turned on  (use the checkbox).

If you don't get it to work you might want to use just one include regex statement like so:

(^query 1$)|(^query 2$)|(^query 3$)....

Good luck... If you don't get to work you might want to post a screenshot of your current filter settings

When linking to another site you pass 'link-juice'. Consequently, you vouch for this website. By adding a nofollow tag to your link, you can avoid this. I'd recommend doing so if you suspect Google might penalize your non-adult website.

You can add a nofollow like so:

example

I've also seen quite a lot of these:

example.com/home.html
example.com/cms/index.html

Hi Brad,

I'd love to have all these loads of quality content for our site. However, there is a limit to what you can use to rank. Suppose you move your blog to your ecommerce domain (which I think is a good idea), you suddenly have a lot of pages on this website. Combine this with the 6000 product and category pages you speak about, and you might just have too many pages to rank properly.

With a site this big you need a perfect internal linking structure:

1. If you spread you linkjuice too thin you might see your rankings drop, I've seen that happen before. Make sure that your authority grows along with your content, and that google can keep up indexing everything.

2. See what pages could potentially rank for high volume and competitive terms, and make sure that they get lot's of linkjuice. Longtail is great but you might want to rank for some high volume keywords as well!

Keep up the good work!

Sven Witteveen

Expand Online

Since creating more than one link to some URL from one page does not increase the linkjuice that is passed to that URL, I recon that Crawlers understand that these are identical links. This isn't easy to verify though...

Although users might not go to the sitemap very often, it is usually a very easy way to make sure some linkjuice is passed to all pages. Especially if the sitemap is linked to from a lot of pages, it usually has quite some juice to pass on. However, the sitemap should never be the only way you link to deeper pages.

Usually creating a 301 redirect in your HTAcces file is an effective way of dealing with duplicate content. You could use the free redirect generator if you're not too familiar with writing htaccess files.

http://www.htaccessredirect.net/index.php

Google Page Speed provides very helpfull information on what factors are import for your site. It determines what is slowing your pages down, and what you can do to fix it. It also distinguishes between various degrees of importance. You can check out the speed test at: https://developers.google.com/pagespeed/.

Good luck!

Google isn't too eager on showing information about links. For example, you could search in google using the 'link:' operand, but the list it will return is far from complete.

'link:www.seomoz.org' returns about 1700 results, but I think we all know they have a quite a few more than 1700 backlinks. Google Webmaster Tools isn't showing a complete overview either, but it's better than using the search engine.

I'd say the best you can do is keep a close watch on analytics and see where traffic is coming from.

Obviously you can use third party tools like opensiteexplorer.org and Majestic for tracking links to a website. Usually Google picks up on links way before they do (in my experience).

Good Luck!

Hi Ben,

I agree with you that some links are not worth redirecting. However, in my experience a dead link never comes alone. Often there is some kind of reason that the link was created, and there might be others you don't know about.

For this reason I usually recommend redirecting all broken links, even if the individual link is not worth the trouble. Obviously there are exceptions to this rule, but most of the time it's worth your trouble.

Sven

I'm pretty sure it doesnt redirect twice. I double checked it with httpFox. It simply redirects only once.

If you want to see for yourself:

(https://addons.mozilla.org/en-US/firefox/addon/httpfox/)

Good luck with your website!

As far as I can see the redirect points to http://earthsaverequipment.com/, so multiple redirects is probably not the issue. Very strange though!?

I do not get a 404 error when i'm going to  http://www.earthsaverequipment.com. Also a redirect is not handled through your robots.txt. This file is used to specify which locations a robot is allowed to enter.

You can use http://web-sniffer.net/ to see what header a URL returns, and you will see that your website returns a 301 redirect, as it should.

The robots.txt file can also be used to specifiy the location of your sitemap, which it currently doesn't. I suggest you add this information.

Recently there was a blog posted on the Youmoz Blog which is worth reading. I think the main idea of this post is that the best strategy for international SEO depends heavily on the company, and the resources they are willing to invest.

The main distintion is made  between:

subfolders of one international domain (.com/us/ .com/au/  etc.):

This is the easiest way to maintain different websites. (low cost, easy to use). However, you will also have the most difficulties ranking the right page in the right country.

seperate country specific domains (.co.uk .au .com)

If there are resources available for maintaining separate websites, building domain authority and creating content for all of them, this would be the best option.

subdomains (au.blindbolt.com etc.)

This is the middle way. Harder to maintain than subfolders but not neccissarily more expensive.

I'd advise you to read the blog carefully and also to study hard on best practices in international SEO, since a lot of people have difficulties ranking the right pages in the right countries.

earnings.png

Well, First I figured the internal links were simply not indexed. This would mean Open Site Explorer would not show info for pages low in your hierarchy. When I checked, it did, so I was initially wrong. However I also noticed a lot of external links from etoyszone.co.uk so I checked that site, only to find myself redirected to your current site.

I think this problem will solve itself over time, since I expect that the deeper pages will not be crawled very often. Therefore I would advise to always double check and use crawler software (like Xenu) to map your internal link structure.

cheers!