Welcome to the Q&A Forum

Hey Spencer, Normally you'd use the meta fragment directive you mention for pages that don't have #! in the URL (see section 3 here: https://developers.google.com/webmasters/ajax-crawling/docs/getting-started) to indicate to crawlers that this site is AJAX. When crawlers account the #! they usually search for the 'crawl friendly' version of that URL which is specified by the 'escaped_fragment' URL parameter. The directive above indicates to crawlers that even though they don't see a hash they are on an AJAX page. The #! approach was an interim method that sites used, which is gradually being replaced by the alternative approach that HTML5 PushState allows. I think if you're still confused the easiest solution would be to get some example URLs for your site (or at least the pattern of the URLs and what markup they have etc., and whether they are indexed). Hope this helps! -Tom

Hey Spencer While I'm not truly sure how this will respond, I bet if you use the structured data tester it will give you an answer. In my research as well, I don't think it's a good idea to use display:none (resource) - but my feeling is if you do this correctly, the breadcrumbs will show in the SERPs. -Dan

Spencer, a most excellent question IMO. I am going to stick my neck out here and say use it on both. It took some thinking and digging for me to wrap my head around it, but I think I have it: You should only use rel=pub on one page of the site. Given that, you use it on the one that is your brand or which has the most traffic, gets lots of plus ones, etc. But, you said this:  We also have a community subdomain... I think the subdomain should be treated as its own site and let it also have rel=pub to the G+ of your main site. The individuals who write, whether for UGC or your experts writing will be rel=author for their pieces. By doing it this way you can aggregate the plus ones to your G+ from all which gives your site a better profile in the eyes of the Google. This opens a lot of doors for you (at G's choice obviously.  Now, I am certainly open to being wrong, but if you think about it for a moment it makes sense in this way: If your firm owns multiple sites (we are a marketing firm and we own a few properties ourselves outside of the main site) but your brand is what owns them, then you can be the pub of all. Again, the individuals writing will be the rel=author.  I look forward to more comments. Most excellent question for forcing us to think.

Hi Spence, IMO you may have two situations: if the old international sites were in other languages you may consider redirecting them to the same language version on the main site if the old domains were focused on specific markets (countries in this sense) you'll better point the 301 the to the specific directory pointing to a specific country (if it was domain.es focusing on spanish users point to domain.com/spain/)

Hi Spencer, The rel="alternate" hreflang="x" tag inclusion along with the Geotargeting option in Google Webmaster Tools (if the websites are not ccTLDs but generic domains) are solutions for these situations: When you have two websites targeting different countries with the same language and they're competing with each other in SERPs, cannibalizing their opportunities and that might end up also with content duplication issues at some point. About your concern: "Currently for some of our keywords US is #1, UK is #4. If we implement rel alternate, will it just remove our US page". Your US website would be removed for the UK search result pages indeed and although in the short term this might seem as counterproductive, the reality is that this result from a relevancy perspective doesn't belong there since you already have your UK website targeting those users especially. This also means that you can also be losing opportunity by having UK users going to the US version that is not as relevant for them, so their experience can be poorer and the conversions might not be the same. Additionally, as I mentioned before you might end-up suffering from content cannibalization / duplication issues with both websites showing the same or very similar content without giving signals to Google that they're not really duplicates but two versions targeting different audiences. Because the previous in the long-run the best is that you make the most out of your international Web versions (country versions in this case) and give the necessary signals to Google so these start ranking in these countries search results instead. In case your US version is a .com domain and it's also attracting rankings, organic traffic and conversions from abroad, excluding the UK (for which you have your own UK website version) then you can leave that website without a specific geolocalization in Google Webmaster Tools and just specifying the language in the hreflang attribute (not the country) but if you have a specific website version for the UK then you should definitely specify that this website (whether it's a .uk ccTLD, subdirectory, subdomain or another generic domain) is geotargeting to this country. I hope this helps!

Thanks guys, We definitely mark up entities that have a chance of showing rich snippets, so far though I haven't seen any of these for purely article markup. http://schema.org/Article I guess that answers my question though, probably not worth the implementation costs at this time.

Hi Spencer I'm honestly not sure if they'll show a 410 (I would assume so). I just check a bunch of profiles in my WMT tools account and didn't see any, although they are less common. This help page and other sources from Google have always seemed to imply they don't think of them as very different, in regards to their impact on search. If your goal is to make sure the pages return the intended codes, maybe you could keep a list and crawl them in list mode in Screaming Frog to check the response codes. If they do show in WMT for you, let us know! I'd be interested to see confirmation of that. -Dan

First I suggest reading this post: http://www.seomoz.org/blog/information-architecture-faceted-navigation-duplicate-content-oh-my You may want to consider limiting the depth of your indexable facets. If these are strictly search pages, and you have an alternate form of navigation, you may consider not allowing any of them to be indexed, as Mike recommended. I couldn't say for sure without knowing more about your particular site and situation. Your bottom line question seems to be "How should we handle these" and in my opinion I think you should focus more on the user than Google. It sounds like a pretty annoying user experience to spend time drilling down five facets deep only to reach a page that has no results. Why even allow them to drill down to that last page? Why not grey out the option? And if they do drill down that far, try showing them some alternatives. If you want to dynamically serve a 404 status code in the http header for such pages that would be fine too, but it could cause problems if you sometimes have products on those pages and sometimes don't since that URL would be constantly coming in and out of the index. A better solution would be to make the page more useful and/or limit the depth of indexation. Also, as Mike alludes to below, if you have another source of navigation on the site the "search result" pages probably shouldn't be indexable at all. It depends on your situation so I'm not going to say for sure that you shouldn't index them.

Also, instruct Google on the way you want them to handles parameters (in webmaster tools).

That is true, but I also have them 301'd to the http version and canonicaled! That is pretty much every possible signal to tell them those pages aren't pages and don't index them. I suppose we can submit the URLs, unfortunately there are a LOT of tag pages. Thanks for the advice Dana!

Likes can impact search results in Bing. But beyond that, they're social proof more than anything else. Likes to your page, however, can be used as a way to connect with your fans over Facebook, and will likely have benefits for Facebook's new Graph Search as well. I would choose building up my Facebook following over having people like my homepage.

You need to point the site to one option or the other, there is no benefit one over the other, but people are used to saying www. In addition to that SEOmoz use www, and for this sort of question I tend to use this site as a rule of thumb guide I hope this helps Sean

Darren is absolutely right. However, it seems unclear as to whether or not you set up 301-redirects from your old URLs to your new URLs. If you did, then the drop should only be temporary as Google figures out where the new URLs are and the old authority is passed to the new pages (or at least most of it). If you didn't do any 301-redirects, and you still have access to a spreadsheet or sitemap of your old URLs I'd encourage you to set up 301 redirects as soon as possible. Otherwise, it will indeed be like starting completely over with a new site, regardless of hyphens or underscores.

That is true. If you nofollow a link, nobody gets the PR.

A good observation Ryan, I tend to agree with you that Google's numbers would increase if more of it's properties were included in the calculation.

Ive seen from £350 for content and design to £1200 for just design (was awesome design tho) S

That certainly works from a functional perspective. Based on the original question and the avoidance of any numbers in the current URL, it seems that would be quite undesirable. Many people want to avoid all numbers in URLs as it looks quite ugly. Including the post id allows a balance between minimizing numerical use, and functionality.

Here is my best guess about what this metric is aiming for: How does the quality vs. quantity of the links to this page compare? If a page has a mT/mR score of around 2, with Trust being 4 and Rank being 2, then you're looking at a content that's linked to from great sources but isn't linked to a lot. If the mT/mR score is .5, with Trust being a 2 and Rank being a 4, then you have a page with a ton of links to it, but few that are from "trusted" sources. My guess is that the latter case might start to look like spam to Google - someone building up low-quality links (blogger comments or even the dreaded linkfarms). I am not sure what to make of a page where the Trust and Rank are both high and the ratio ends up looking close to 1. We have one page that falls under this category - the Trust and Rank are both higher than the competing Wikipedia entry, but the mT/mR ratio is lower because Wikipedia has a lot less strong content and fewer strong links directly to that page on an admittedly niche term.

Same here...Started one yesterday and still not complete.