Welcome to the Q&A Forum

Any feedback from your web logs on what Google's crawling?

Hi there,

This is definitely a crazy problem! It looks like you've done what you should, but Google's ignoring you.

Here's a theory, though: I don't think that Google loves the idea of there being no "home" page; it probably only expects domain.com/home or domain.com/default.asp or domain.com/index.html as alternatives to domain.com, so seeing http://weekend.visitsweden.com/ redirect to http://weekend.visitsweden.com/no/ could be confusing it.

Is there a reason why you don't want http://weekend.visitsweden.com/ to be the homepage?

Kristina

Hi there,

Have you run a crawl on your site to see if there are a lot of links pointing to the old URLs? If Google sees more links point to the old version of the URLs rather than the new version, it's possible that it thinks that the old pages aren't really gone for good.

• Kristina

Hi Joe,

There haven't been any major image algorithm updates since the beginning of August that I'm aware of. Have you looked at the queries that you rank on image search for to see where specifically you're dropping? There could be another site that popped up in early August that's outranking you.

Best,

Kristina

You have the rel="alternate" tags from the page back to the mobile version, right? That seems odd...has anyone else had this with mobile and desktop pages in WMT?

This is a pretty thorough outline of what you need to do: http://moz.com/blog/web-site-migration-guide-tips-for-seos

My steps are usually:

1. Identify pages that get significant organic traffic by pulling the Organic Traffic report in Google Analytics for the past year or so.
2. Identify pages that have a significant number of links (or, have links from high traffic sources) in Open Site Explorer.
3. Map where that content should be now, and 301 redirect to new pages.
4. Completely remove all old pages from the index by 404ing them and making sure that no links on new pages point to old pages.

Sounds quick and simple, but this definitely takes time. Good luck!

There really isn't a problem with having 100,000 404 "errors." Google's telling you that it thinks 100,000 pages exist, but when it tries to find them, it's getting a 404 code. That's fine: 404s tell Google that a page doesn't exist and to remove the page from Google's index. That's what we want.

The real problem is with your site migration, as FCBM pointed out. If you properly 301 redirect old pages to new, Google will be redirected to the new page, it won't just hit a 404. If you fix the problems with the site migration (not focusing on Google too much), the 404 errors will naturally subside.

The other option is to just take the hit from the migration, and Google will eventually remove all of these pages from its index and stop reporting on them, as long as there aren't live links pointing to the removed pages.

Good luck!

Yup, you've got it!

Hi Critical Mass,

Before I answer your direct questions, I think you're putting yourself in a tricky situation by creating vanity 301 redirects to those dynamic mobile URLs. If someone ever links to the mobile version of your page, they're going to use the URL with parameters, because that's the page they end up on. That means that all inbound links will point Google to your parameter URLs and all internal links will point to the static URLs you've created. Link equity will be split, and all pages will suffer for it.

It's true that Google understands static URLs a bit better than it understands URLs built with parameters, but it does understand that sometimes parameters define content. I recommend getting rid of those static URLs, then using Google Webmaster Tools to explicitly say, "these parameters define content." You can learn more about how to do this here: https://support.google.com/webmasters/answer/1235687?hl=en

Now, to answer your questions:

1. You should only include the URLs that you want Google to index. If you follow my recommendation above, this is now an easy question to answer.

2. Yes, use the rel="alternate" tag on your desktop pages! Like I said, Google understands that parameters can determine content. You want to connect the two pages as much as possible, even if you can't canonical back.

Hope this helps!

Kristina

I wouldn't use rel="canonical" here. It sounds like the pages do look similar, but if they're advertising different events, they're not really the same thing, and canonicaling them to each other could look spammy.

Instead, I would make it as clear as possible that Googlebot sees unique information on each page:

• Definitely have different default content on the homepage, so Google sees that your homepage is unique from your events pages. I'm not sure exactly how you're switching out content by IP address, but use the Fetch as Googlebot feature in Google Webmaster Tools to see what Google is seeing on your homepage.

• Use unique information about events and location in the page title, meta description, and headers of these events pages. Ex:

• www.townscript.com/mumbai

• **Page Title: **Events in Mumbai | Townscript

• **Meta Description: **[Write something unique for each city...yes, this will take time, but it will be worth it]

• **H1: **Trending Events in Mumbai (this is currently an h3)

Good luck!

Kristina

No need to worry about hreflang (and therefore x-default) unless you're working with international versions of your sites! If Google doesn't see anything referencing international, it will figure out the language and country from the ccTLD and the language you're using on your site. Hreflang tells Google where the international versions of a page are, it isn't necessary to designate a language and location.

Best,

Kristina

You only really need to use hreflang in either the XML sitemaps OR on the page: not both. Google's pretty smart!

For more detailed info about how to optimize for international SEO, I like to use Kate's tool: https://www.distilled.net/blog/seo/choose-your-international-strategy-tool/ 

Best,

Kristina

The key here is that you should use hreflang when you have two pages with exactly the same content, but in 2 different languages. If those different languages are on different subdomains or ccTLDs, the rule still stands.

But, if you have different content for different countries, just use the ccTLDs without hreflang.

For future reference, whenever I have questions about international SEO, I turn to Kate Morris, who wrote this helpful post: https://www.distilled.net/blog/seo/choose-your-international-strategy-tool/

Thanks for reaching out to our community: it's always great having new Mozzers! Let me know if you have follow up questions.

Best,

Kristina

Hi Ash,

When Google stops indexing a page, it has to find it again and see that it's no longer "noindex"ed. Submitting it to GWT often works, but sometimes you just have to wait for the crawler to find your page again. I looked at the cache dates for other similarly deep pages on your site, and it looks like Google only crawled them a week ago. Deeper pages were crawled a week and a half ago. So, you should probably give yourself a couple of weeks to allow Google's crawler to find the page before worrying that it's a technical issue.

Good luck!

Kristina

I've heard anywhere from 2 - 3 weeks to 2 - 3 months.

Remember, though, that connecting pages with an hreflang tag doesn't make the /ireland/ section of the site magically rank as well as the UK part of the site does in the UK. It's a different page and Google will treat it as such, although Google should definitely show the right page for branded searches.

To increase rankings for the /ireland/ section of the site, you need to prove that your Ireland section is as valuable to Irish people as the UK section is to British people. Do some promotions to build links and get some social media buzz going: that'll help you the most in the long run.

Good luck!

Yes, this will be totally fine. Google ignores bookmarks (#) like this, so Google will just see a redirect from http://www.mysite.com/shoes/red/description to http://www.mysite.com/shoes/red#desc.

Good luck!

Kristina

Another good idea that I often use: where are your customers pain points, and what sort of information might they be looking for online? Use Google Keyword Planner to choose a subject that you think your customers are interested in, but that there are no good answers on. Research the answer, write it up in a guide, and put it on your site. It'll build your brand awareness even when people aren't actively looking for your services, plus any inbound links it gets will help your domain authority.

Hope that helps!

First of all, I'd like a little more information about the "legal reasons" that are forcing to you create ccTLDs for Canada and the UK. What you're doing now is actually ideal for international SEO when it comes to the same language, so I wouldn't want to change things unless it's really necessary.

• Is it because the EU has stricter laws regarding privacy? In that case, I'd recommend following the strictest laws for the entire .com site and leaving it at that.
• Is it because your company has to offer different legal information from country to country? If that's true, I'd post all legal information on my site, for all countries, and let visitors look at the pieces that are most relevant.
• Is it because you're starting to sell different products? In that case, yes, you probably need a new ccTLD since you technically have different sites, but that means that you can't really use hreflang, since your product pages will be different.

If it's something I didn't list, and you absolutely have to create new ccTLDs, the best recommendation is to use hreflang. You can either use it in the of your page, like you said, or in the XML sitemap so the extra code doesn't have to be loaded with each page.

Hreflangs work a little like canonicals, so Google can choose to pay attention to the international versions of the site or ignore them, but eventually yes, your UK and Canadian sites should rank as well as your .com site. There will definitely be a dip in traffic as Google figures things out, though.

Hope this helps, and good luck!

Hey Djacko: Has Google figured it out? Is the right version of your page indexed now?

I took a look, and I can tell you this: Google definitely knows that you have a mobile site and can find your mobile pages.

I wanted to look into why you were having trouble ranking, but since I don't have access to your Google Webmaster Tools account, I used SearchMetrics to find your top keywords. Looks like the strongest keyword you have is "veterinary websites" - is that accurate?

In the case of this particular keyword, you're ranking #2 on desktop but #4 or #5 on mobile. The reason is, the page that ranks well doesn't have a mobile version. Google is still listing the desktop version, but understands that when searchers click it, they'll get redirected to your mobile homepage, which Google doesn't see as as good of a match. So, your mobile site has been pushed down a bit.

I don't know if this is the case for the other rankings drops you've seen, but you might want to check if this is the issue. If it is, creating a mobile version of the page that's ranking should help increase its mobile rankings.

Hope this helps!

Kristina