Welcome to the Q&A Forum
Browse the forum for helpful insights and fresh discussions about all things SEO.
In the new blog post here, she links to mobilemoxie, a site which allows you to do searches on various different smartphones. I tried a search for my brand, "ixl.com" on an iPhone 4 and a Android phone, the Motorola Droid X. On the iPhone, our brand appears in #1 like normal, but on the Android phone, our result doesn't appear on the first page.
Any ideas as to why this would be the case? We're currently developing a mobile version of our site but we don't have one yet. Could that be the issue?
Hey all!
Does anyone have any experience with the following ad management platforms (or any other ones they'd like to weigh in on)? I'm going to be using one in-house, not as an agency.
Have you had good experiences with Boost CTR? We're looking into using them, but I'd love to hear what you have to say, good or bad.
On one site I work on all of the content is loaded with AJAX. I've seen evidence that Google is indexing this text, because if I search for this AJAX content, Google is returning those pages on my site for which it appears on. There aren't many backlinks to these pages, and none that I've found with those phrases as anchor text. So if Google is finding this text out from somewhere else, I can't figure it out.
I'd still rather have our content loaded normally, but I wasn't given a choice, so I'm glad to see it's working. I'd think long and hard before setting up HTML snapshots as given here: https://developers.google.com/webmasters/ajax-crawling/, as doing so would be a considerable amount of work, especially considering how far Google has already come with indexing AJAX content.
Alan is right. Your question reminded me of this relevant WBF. Enjoy! http://www.seomoz.org/blog/title-tags-is-70-characters-the-best-practice-whiteboard-friday
Google doesn't care, but Bing may not like this. Read http://nickroshon.com/seo/google-bing-disagree-on-relcanonical-implementation for more info.
The bots won't care about that. If you have your site on www.company.com, your robots.txt will reside at www.company.com/robots.txt, and its directives will apply to any pages living under www.company.com. When a bot comes to www.company.com/blog, it'll look for the robots.txt at www.company.com/robots.txt to see if it's allowed to crawl there. It won't look in a subdirectory. Robots.txt always resides on the root level.
If you had your blog at blog.company.com instead of company.com/blog, then you would have to have a separate robots.txt at blog.company.com/robots.txt. As you have your blog in a subdirectory rather than a subdomain, one robots.txt is all you need.
Disallowing in your robots.txt keeps the bots from indexing your pages going forward, but Google may keep returning them in search results. This post has great explanations about ways to remove pages from indices: http://www.seomoz.org/blog/robot-access-indexation-restriction-techniques-avoiding-conflicts
The surefire way to get them out of the index is to remove the disallow from your robots.txt, and add a meta noindex tags on all the pages you want removed. Once they're reindexed by Google, they'll no longer appear in SERPs.
Looking at http://googlewebmastercentral.blogspot.com/2011/12/new-markup-for-multilingual-content.html and http://support.google.com/webmasters/bin/answer.py?hl=en&answer=189077, you would only want to implement rel=alternate tags when the content is duplicate content. So if you write new content in a different language on a subdomain that doesn't appear on your main domain in a different language, don't specify a rel=alternate. If it's a translation of content on your main domain, go ahead and add it.
The difference in the implementation when you have pages with completely different languages vs. small regional differences is if the pages are in completely different languages, you won't want to implement canonical tags pointing to one of the variations.
Re-reading through those articles, it's not 100% clear to me what adding this does when you don't implement the canonical tags. Google says "This markup tells Google's algorithm to consider all of these pages as alternate versions of each other."
Matt Cutts said at one point that page speed effects less than 1% of searches. You can read more here: http://www.mattcutts.com/blog/site-speed/.
So, I wouldn't expect it to help much. But think about how much happier your users will be! That's where the real gains are. Over time, it might help you a bit in the SERPs because people should bounce less frequently as a function of how fast your pages load.
From what I've read, they're indexing Facebook and Disqus comments on blogs. On the site I work on, on certain pages we load the main page contents via AJAX, and Google is reading and indexing that content.
I'd be wary of taking the time to put together HTML snapshots, as Google is getting better and better at this. You could set up a prototype of your AJAX solution and wait and see if Googlebot can index the AJAX loaded content.
Make sure you're setting your location for your domain/subdomains/directories correctly in Google Webmaster Tools.
For Bing, you'll want to set the language and location targeting on your page, via meta tags or HTTP headers: http://www.bing.com/community/site_blogs/b/webmaster/archive/2011/03/01/how-to-tell-bing-your-website-s-country-and-language.aspx.
If you're going to have duplicate content across multiple languages, be sure to read: http://googlewebmastercentral.blogspot.com/2011/12/new-markup-for-multilingual-content.html.
I remember reading somewhere that the title attributes have no weight for ranking purposes. If you add a title to a link, it'll add a tool tip showing that text when a user hovers over the link. So if you think your users would like that, go ahead and add it.
Google can index and render both CSS and JavaScript, and has an uncanny ability to sort out bad intentions, so if you're trying to mask links with these, Google will (eventually) figure it out. Your developer's assumption is wrong. Google has come a long with with this, and can even index some AJAX content (when JavaScript loads additional content to a page from the server).
Lots of navigation use accordions and other widgets that will display or hide navigation links depending on the page state. That's fine. Just don't go out of your way to hide links.
You used to be able to do this, but PageRank sculpting is no longer valid. I wouldn't nofollow any internal links. Nofollowing links burns the PageRank rather than sculpting it. Here's an example:
Page A links to Page B, C and D. Page A has 6 link juice "votes" to cast to these pages it links to. Page A passes 2 votes to each page, B, C and D.
You nofollow the link to page D. Now page A passes 2 votes to page B and C, and no votes to page D. 2 votes are lost by the nofollow. It doesn't pass 3 votes each to pages B and C.
My instincts say I'd rather exceed the 100 links than trying to get the count down by nofollowing footer links, but I've never had this issue before so I can't say for sure. Google may follow nofollowed links, so I have a feeling this doesn't really get the link count down in Googlebots eyes.
Here's Matt Cutts take on it: http://www.youtube.com/watch?v=r1lVPrYoBkA.
He says Googlebot will follow 301 redirect chains. Once you get to 4 or 5 in a row... he says Googlebot might bow out. He recommends 1 or 2, or maybe 3 in a row.
So based on what he said, you should be completely safe doing 2 redirects.
My advice is to not nofollow any internal links. It burns your link juice; it no longer sculpts it. If you really don't want the page it's going out to indexed, either exclude it in your robots.txt, or even better, stick a meta noindex tag on it.
You can't really avoid doing the math to figure out the probability that the better ad is actually besting the older ad. It's a function of both the impressions and clicks so it's not easy to eyeball. I threw together a little spreadsheet that should do the math for you if you put in the click and impression data. I checked it with some data and I think it's working. 
It'll calculate whether we're 90% sure, and 95% sure which is better. Just put in your ad numbers and you should be good to go! Usually, if the 90% variation matches up with my expectations, I go with it. Otherwise, you might wait for 95% just to be sure.
My worksheet: http://www.mediafire.com/?pmu2mc6n5f6vw5s
A/B testing people deal with confidence intervals all the time, so they know their stuff. Here's some reading about it if you're interested:
Hope this helps!