Welcome to the Q&A Forum

Yeah, I'm with Alan - what works in theory and what works in practice can really vary with the site - if you do both at once and something goes wrong, you won't know which tactic was the problem. Even if it goes right, you won't know which one worked.

Keep in mind, too, that 301s and canonicals have different purposes. They may act similarly for search, in some cases, but they're very different for site visitors. A 301-redirect will actually take the visitor to a new page. A canonical will let the duplicate URL be visited but remove it from search results. In most cases, one is more appropriate to your situation than the other.

It's always hard to speak in generalities, but my gut reaction is that Alan's right - if the canonical tags were implemented properly, having your rankings tank from this kind of implementation seems very unlikely. A couple of possibilities:

(1) Are your canonical URLs being used in internal links? If you tell Google that one version is canonical but then act as if another version is canonical, it can cause problems.

(2) Are you sending any other, conflicting cues, like 301-redirects or Webmaster Tools parameter handling?

(3) Is it possible that your canonicalization was too broad? In other words, did you end up de-indexing some product variations that were driving long-tail traffic? For example, let's say you had a product in red, blue and green and you canonicalized them all to the "root" product page. In theory, that might be a good thing, but if people were searching for specifics and you had a lot of long-tail rankings ("buy product in red"), then it could be bad.

Oh, sorry, it's a session ID, not a tracking/affiliate sort of ID. Honestly, the best solution is to avoid URL-based session IDs entirely, and store it in a cookie or session variable, but yeah, I realize that's not always feasible.

In this case, the 302-redirect should help keep link-juice at the root URL, and is probably a good bet. I think adding the canonical tag to the parameterized versions is a good backup, though. You could also block that parameter in Google Webmaster Tools, since it really has no search value at all.

So, it's something like this?

www.example.com?id=12345

(1) canonical to -> www.example.com

(2) 302-redirect to -> www.example.com

Is the 302 intended so that visitors don't bookmark the ID'ed version? The problem is that the 302 is essentially telling Google to leave link-juice at the ID'ed URL, while the canonical is telling Google to consolidate link-juice to the root URL. I think I get your intent, but it's a mixed signal to the search engines. In this case, I do think that a 301 is the way to go, unless I'm misunderstanding.

When you say "delete", do you mean delete it from the home button (and just use the root version), or delete the page? Keep in mind that the "index.html" page also drives the root version of the site ("www.example.com/"), so you definitely don't want to delete that file.

Whether you use 301-redirects or canonical tags (either should be ok, but it sounds like the canonical tag would be easier for you to implement), it is important to use a consistent path. So, you do need to change the button link to match the canonical version of the URL. Don't do this until after you've got either the 301 or canonical tag in place.

Remember, if you put this in the PHP code, it has to be conditional - you don't want to redirect your canonical version unnecessarily. Since your home-page is just one template, you'd be firing a 301 for every situation.

(1) Honestly, I tend not to double-up, if for no other reason that you can't really tell what's work and what isn't. Keep in mind, too, that these tools do have different purposes. 301-redirects impact everyone (users and bots), whereas rel-canonical is only for search. If a 301 is appropriate, then just use a 301.

(2) If the address bar isn't changing, your 301-redirect isn't working. Test it with a header checker:

http://tools.seobook.com/server-header-checker/

A rel-canonical tag will not override browser behavior (at least, not at this point in time).

(3) It can take weeks to clear, and it sounds like your 301 isn't working right, so that's going to exacerbate the problem. The page has to re-crawl and re-cache, and GWT may still show the message for a couple of weeks after that.

Personally, I'd drop the canonical and fix the 301-redirect.

Canonicalization is a bit of an "it depends" answer in a lot of cases. If the domains are really identical and if you want to discontinue using the two duplicate domains, then 301-redirects are probably the best long-term bet.

If you need to continue using the other two domains (as landing pages for other marketing campaigns or paid search, for example), then 301s aren't going to be viable, and cross-domain canonicals should be a decent choice. I'd also say that if you just have a small number of duplicates pages (not the entire sites), then canonical tags are a good bet.

The only time I'd be cautious with 301s is if: (1) one of the domains has a bad history or questionable link profile, or (2) you're 301'ing a large number of domains - two domains is probably not an issue. In that cases, canonical tags may be a little safer (people aren't gaming them as much as 301s - yet - so Google isn't quite as suspicious).

I think Sven's essentially correct, although I suspect the word "Shimano" is what we're flagging. You've got the main Shimano page that splits out into sub-pages, and every link for the sub-pages has "Shimano" in the anchor text. The problem is that it IS keyword cannibalization, in a sense, and yet it also makes perfect sense in this situation.

I think this is pretty common in search pages, and probably isn't a big issue. Honestly, internal search pages and Google don't always get along in general, so it can be tricky. You could draw "Shimano" out into a header and then not use it in every piece of anchor text (it could start to look excessive), but practically speaking, I doubt it's going to have much impact either way.

I'd agree but probably feel a bit more strongly about it - post-Panda, these kinds of pages can definitely cause you harm (depending on the scope). If you're not linking to them and they're off on a sub-domain, they may not be creating any practical issues, but it's still better to prevent problems. META NOINDEX probably is a good bet here for prevention.

Yeah, your best bet is probably to switch the canonical to the old/main page "/widgets.php" - just removing the canonical tag doesn't always get Google's attention (or, at best, takes quite a while). If you can designate a new canonical, that should speed things up.

Our on-page report may just be looking at the longer version with more content as being a bit better, but it's really tough to say. Without knowing the situation, I'm not clear on how the 2 pages differ or what kind of pagination you were using.

Does your site use the "index.php" version of the page with parameters to handle logins? In that case, you can't 301-redirect the page across the board. You'd have to do it selectively (only for the version with no parameters), and that's going to take custom-coding. You'd probably be better off with a canonical tag, in that situation - it would keep Google from indexing the variants (including the login pages, which have no search value).

Good additional info on Penguin from Danny Sullivan today:

http://searchengineland.com/google-talks-penguin-update-recover-negative-seo-120463

Looks like it is on an "offline" data update, like Panda. So, recovery could be delayed in many cases.

For everyone's privacy/security, we'd prefer you not share emails and account info via Public Q&A. Please use the private messaging system. Thanks.

I'm only seeing Google index the "www" versions at the moment, except for 1 or 2 exceptions. When you used full paths, were they identical to your canonical URLs? In other words, if Google is indexing the "www" version (which they are), the full path should be "www." Otherwise, you'd be potentially creating internal duplicates.

Ultimately, you're only get links from one additional site (the number of links is far less important than your linking root domains), and it's very likely Google will see the connection and not value those links much - they won't penalize it, but they may turn down the volume. Meanwhile, you're basically splitting your back-links, social presence, and promotional efforts. Any links to the external blog are all getting funneled into one linking domain. If you pointed them directly to your main site, you could actually have more links, in Google's eyes. I can't say that there's never a situation where I would house a blog externally, but it's rarely a plus for SEO.