Welcome to the Q&A Forum

I have a post on the subject here - it's very long, because it's a complex subject:

http://www.seomoz.org/blog/duplicate-content-in-a-post-panda-world

We're not saying this is definitely the problem - just that you should be aware of how complicated it can get. Unfortunately, it's hard to tell without really looking at the site. A lot can happen to hurt a site's rankings, and the EMD update was just one piece of the puzzle.

Keep in mind, too, that a lot of duplicate content is accidental. Google doesn't care about pages, per se - they care about unique URLs. So, if you have 300 unique pages, but something about your CMS translates that into 5,000 crawlable URLs, then you could definitely have problems.

Unfortunately, we don't currently track rel=prev/next - definitely on the list for next steps. As best I can tell, your rel=prev/next tags are properly implemented, and they are a good bet for paginated search results like this.

There is one issue, though - your category pages within the blog have the same title as the blog home-page. It would really be a good idea to make those unique. Otherwise, they could slightly cannibalize each other.

I don't want to give examples, sine this is a public question, but hopefully you know which pages I mean.

If this is really related to the EMD update, then I'd agree with Charles - starting over is a bad idea. My best guess is that the EMD update wasn't a penalty, per se - it was more like Google lowered the volume on EMDs. In other words, having one isn't bad now - it's just that it's not as good as it used to be. There's no way to fix that really (you can't turn the volume back up), but the risks of switching domains would probably far outweigh the benefits.

To back up Marie, though, a lot happened right around the EMD update, and it's really tough to diagnose. I'd definitely look at Panda factors, like "thin" content. Try to look at the site from Google's POV - what you view as unique doesn't matter, frankly. You could be spinning out URL-based duplicates, for example, and not realize it - that's more of a technical SEO issue (you're not doing anything devious, but the site may still be giving Google problems).

The other issue to consider is whether your EMD has caused you to really pile on exact-match anchor text, especially keyword-loaded anchor text. This could trigger Penguin or similar problems. This is often correlated with EMDs, even though it wouldn't necessarily be a result of the EMD update.

If they've really just turned down the "volume", then you have to get other ranking factors in play - build more relevant, authoritative links, increase your social signals, etc. In other words, focus on aspects of SEO beyond simple on-page ranking factors.

I think Andy's absolutely right - I've seen too many situations where mixed signals caused crawl/index and even ranking problems. Ultimately, the canonical URL should be canonical in practice and used consistently. Otherwise the canonical tag is just a band-aid.

The other problem is that you naturally end up attracting links to your non-canonical URLs, because those are what people can see. Long-term, that compounds the situation.

Now, is it catastrophic? Unfortunately, that's really tough to say. I've seen situations where Google honored the canonical tag even without internal links and the site was ok. I just think it's a significant, unnecessary risk. Unfortunately, like Andy, I don't know of any clear documentation on the subject.

I suspect this isn't so much a general SEO problem as a local SEO problem. Those two worlds have really diverged a lot these days. Generally, you're ranking well for brand terms (you even have expanded site-links) and you come up first in the local 7-pack for "insurance newberry florida".

I suspect the problem is related to Gainesville, and that your local cues aren't very strong. The location (not being in Gainesville) is a stumbling block, and there's no easy answer for that. There are other cues you can tap into, though - the big one is reviews. Reviews still have a big impact on local rankings, and legitimate reviews can go a long way (don't spam them).

David Mihm's piece on Local Ranking Factors is still the go-to, and it shows just how complex this has gotten:

http://www.davidmihm.com/local-search-ranking-factors.shtml

The main thing is that on-page factors (like Gainesville-related keywords) will only take you so far these days. Google is going to treat any "Gainesville"+"insurance" query as having local intent.

Are your sites cross-linked at all (I'm not seeing anything obvious)? The cascade effect across all of them almost suggests they were either (1) seen as a link network, or (2) you're using the same link strategy across all of them.

I'd watch the keyword stuffing - your titles are pretty aggressive, and then you repeat those lists in the copy at the top and bottom of the page. It's borderline at best. I don't think it's enough to get multiple sites penalized, but it may not be helping you right now.

You've got a duplicate of the London Escape home-page floating out there in the index:

http://www.londonescape.com/index.php?option=com_wesearch&task=loginVIAFacebook

Might want to add rel=canonical to the home-page, just to sweep up things like that.

You're indexed and I'm able to get the site ranking on very targeted keywords (exact-match and fairly unique), but you fall off the map for everything else. That's definitely indicative of a link-based penalty.

You have to be very careful about this, as Keri suggested. Google can crawl some JS, especially basic stuff, and if they detect a clear intent to present different content to crawlers and people, you can get into trouble fast.

I would say, though, that pop-overs are pretty common these days. These are usually just a CSS-styled box that appears, with an ad, survey, etc., often the first time you visit a site. Even some reputable survey engines used them.

Now, I still think you should absolutely test this for usability issues, but if the pop-over is just an overlay that appears the first time a user visits, the SEO consequences should be minimal.

Definitely agree on A/B testing. Nothing will win the argument faster than definitive data that this approach is costing sales.

Thumbing up both answers - I think they've got you covered. This is definitely a situation where you should try to sort out why the deeper page is ranking. It could be a positive that you should try to encourage (disrupting that could harm your ranking, ultimately) or it could signal something about your home-page that needs work.

Rand had a good post a while back on the subject:

http://www.seomoz.org/blog/wrong-page-ranking-in-the-results-6-common-causes-5-solutions

Just wanted to follow up for general reference (omitting details of the site) - the OSE counts can be a bit confusing. The top count never actually changes based on the filters - it's the unfiltered, "raw" count, for lack of a better term. I believe the download pulls only Followed/301'ed links (under the "Show" menu) - in other words, links that potentially pass PageRank.

If you want to DM me the domain (or email peter@seomoz.org), I'll take a look. There are a lot of factors that can impact the count, so it's really tough to say without seeing what you're looking at. Total links can sometimes explode based on a single site-wide link (even from one domain), for example.

I think it can get pretty complicated, but a couple of observations:

(1) In my experience, NOINDEX does work - indexation is what Google cares about primarily. Eventually, you do need to trim the crawl paths, XML sitemaps, etc., but often it's best to wait until the content is de-indexed.

(2) From an SEO perspective (temporarily ignoring Panda), a 301 consolidates link juice - so, if a page has incoming links or traffic, that's generally the best way to go. If the page really has no value at all for search, either a 404 or NOINDEX should be ok (strictly from an SEO perspective). If the page is part of a path, then NOINDEX,FOLLOW could preserve the flow of link juice, whereas a 404 might cut it off (not to that page, but to the rest of the site and deeper pages).

(3) From a user perspective, 301, 404, and NOINDEX are very different. A 301 is a good alternative to pass someone to a more relevant or more current page (and replace an expired one), for example. If the page really has no value at all, then I think a 404 is better than NOINDEX, just in principle. A NOINDEX leaves the page lingering around, and sometimes it's better to trim your content completely.

So, the trick is balancing (2) and (3), and that's often not a one-sized fits all solution. In other words, some groups of pages may have different needs than others.

Agreed - my experience is that NOINDEX definitely can have a positive impact on index dilution and even Panda-level problems. Google is mostly interested in index removal.

Of course, you still need to fix internal link structures that might be causing bad URLs to roll out. Even a 404 doesn't remove a crawl path, and tons of them can cause crawler fatigue.

We strongly suspect that canonical tags lose a portion of link "juice" just like 301s do. Otherwise, they could be abused.