301 or canonical for multiple homepage versions?

[kimmiedawn]

And I just did a site operator search fror default.asp and index.htm and they are no longer in the index, though I never removed them after the 301 redirect. So I'm assuming Google did the math as expected.

[ThompsonPaul]

Sorry, I have to categorically disagree with Sangeetac here. Kimberley, your understanding is entirely correct.

There's so much misinformation and misunderstanding about canonicalization, and there doesn't really need to be.

The rel canonical tag is ONLY to be used when there is a justifiable reason why the multiple duplicate pages in question should actually be reachable via different URLs. Unless such a reason exists, you should ALWAYS use a 301 redirect.

Put another way, duplicate page issues should ALWAYS be resolved using 301 redirects unless doing so would harm the user experience.

In the case of multiple versions of a home page, there is absolutely NO BENEFIT to the user to be able to reach that page via multiple different URLs. (In fact the multiple URLs could be confusing.) Therefore, the 301 redirect should be used to FORCE the user (and search engine) to the single version the site owner has decided should be the primary URL. Again, the other URLs are pointless, so the user should be blocked from ever being able to reach them. Using the 301 in this case has the added benefit of explicitly and automatically telling the search engines that all the redirected URLs for that page should be dropped from the index as they are non-functional. (And that's exactly what happened, Kimberley, as you saw default.asp and index.htm disappear from the index.)

The ONLY time you'd use a rel canonical tag in this case, instead of a 301 redirect, is if for some reason the site owner doesn't have sufficient access to the sever files to be able to implement a 301 redirect properly. In that case, a rel canonical tag can be used as a "better than nothing" solution.

The classic example of when to use a rel canonical tag is in the case of a page that lists a large number of products. That default (and therefore canonical version) version of the page may very well list the products in the order they were added to inventory. There may well be additional versions of the page, each at a different URL (often using query parameters, for example) which lists exactly the same content, but one page sorted by size, another version sorted by colour, yet another sorted by price.

Obviously in this case, 301-redirecting those additional dupe pages to the original page would be harmful to the user experience, because the user would never be able to access the content sorted the way they want to see it. So each duplicate version of the page has value to the user and therefore MUST remain available to the user, and therefore to the search engine.

In this case, the site owner adds a rel canonical tag to each of the dupe pages pointing to the primary (canonical) version, to make it clear to the search engine that the dupe pages are intentional secondary variations that should pass all their ranking value back to the primary page. But NOTE! Google themselves say they consider the rel canonical tag as a suggestion only, and will ignore it if they deem it incorrect or manipulative. So it is NOT NEARLY as strong an indicator as a 301 redirect which forces both user and search engine into the desired behaviour.

Sorry if that got a little long-winded, but once we get clear on the basic purpose of 301 redirects vs rel canonical tags, it becomes much more straightforward which to use when. The confusion usually stems from folks trying to justify using canonical tags for purposes for which they were never designed. They are not a panacea or a catch-all for fixing site architecture mistakes.

Hope that clears, rather than muddies the waters?

Paul

[AlanMosley]

Agree with Paul here.

A 301 is a directive to the crawler, while a canonical tag is only a hint and is not always followed. Bing for one will ignore canonical tags if it believes they are misused.

as for the mention of "multiple 301 redirect" .

You do not need to have a 301 redirect for every url, just follow the logic
if HTTP_Host is not myPreferredDomain then redirect to myPreferredDomain

[kimmiedawn]

Thank you Paul! Knowing the WHY behind something always solidifies my understanding of a concept. It does make me wonder why the cheat sheet linked by Sangeetac and Moz's Beginner's Guide to SEO both use that scenario as an example for a 301 redirect - it certainly adds to the confusion.

Alan - while we are on the subject...here is the code I used for the 301 redirect, which was my first ever using .htaccess. The problem is it redirects twice from certain starting points: first from site.com/default.asp to www.site.com/default.asp then to www.site.com/.

RewriteEngine On
RewriteCond %{THE_REQUEST} /index.html? [NC]
RewriteRule ^(./)?index.html?$ /$1 [R=301,L]
RewriteCond %{THE_REQUEST} /default.asp? [NC]
RewriteRule ^(./)?default.asp?$ /$1 [R=301,L]
RewriteCond %{HTTP_HOST} !^www..* [NC]
RewriteRule ^(.*) http://www.%{HTTP_HOST}/$1 [R=301,L]

Initially I tried this (which I think is what you are suggesting?), but if I remember correctly, it wasn't taking care of default.asp or index.html (don't quote me on that):

RewriteEngine On

RewriteCond %{HTTP_HOST} !^www..* [NC]

RewriteRule ^(.*) http://www.%{HTTP_HOST}/$1 [R=301,L]

Thanks for helping me do things the right way!

[kimmiedawn]

I would like to politely add (and first say that I do appreciate anyone's attempt at a helpful response) that it's a very delicate matter to be giving answers without 100% certainty or valid first-hand experience. At least we as a community should be careful to include somewhat of a disclaimer should our answers be anything less than certain. I, for one, do not want to steer anyone down a dangerous path. I do understand it's my responsibility to gauge the accuracy of an answer before I implement it. Think of the potential consequences to someone else if they implement bad/incorrect advice on a site - ouch!

[AlanMosley]

you are correct, it want do the index.html or default.asp

but this rule will solve all your domain problems, not the "!" mean not, so if not the desired domain, then redirect to the desired domain not matter what the domain is. this will fix non www, or any other secondary domain you may have such as oldDomain.com or mergedSite.com

Options +FollowSymLinks RewriteEngine On
RewriteCond %{HTTP_HOST} !^www.domain.com$ [NC]
RewriteRule ^(.*)$ http://www.domain.com/$1 [R=301,L]

but if you have a page called default.asp. your site is a classic ASP site, ASP is a Microsoft technology and would be on a Microsoft web server IIS. if so 301 redirects are very easy to do.

Is your site ASP.is it on a Microsoft IIS server?

[kimmiedawn]

I didn't build the site, nor do I have anything to do with the hosting, but I do have FTP access. I can't remember how or why but I came to the conclusion that it's ASP Classic. Is there a quick way for me to double check? All the files end in .asp if that helps.

[AlanMosley]

Yes that's classic asp

what sort of server is it on windows?

What sort of webserver IIS? if so what version, if it is 7 or greater it is very easy to do your redirects if you have access to the controlpanel

[kimmiedawn]

So you are saying I should use the following exact code instead of the code I have:

RewriteEngine On
RewriteCond %{HTTP_HOST} !^www.domain.com$ [NC]
RewriteRule ^(.*)$ http://www.domain.com/$1 [R=301,L]

And that will take care of the default.asp and index.html as well? And it won't redirect to the www in between, thus creating 2 redirects, correct?

If you don't mind explaining, what's the working difference between the first code I tried (below) and the above code that makes the above code work?

RewriteEngine On
RewriteCond %{HTTP_HOST} !^www..* [NC]
RewriteRule ^(.*) http://www.%{HTTP_HOST}/$1 [R=301,L]

[kimmiedawn]

I don't have access to the control panel and the person who does never did the redirect when I asked so i just did it myself with .htaccess. So did you say I should change it to your code and that will handle everything, including the double redirect?

(Edited out a question that I figured out the answer to.)

[kimmiedawn]

Could the code I used explain why Open Site Explorer won't let me view link data for www.site.com, but shows me www.site.com/default.asp instead? And within that data it shows an incoming link from www.site.com as a 301 redirect. But when I check it in a browser or redirect checker, it shows www.site.com as the final destination. My head hurts.

Edit: (This is what OSE says "The URL you've entered redirects to another URL. We're showing results for www.site.com/default.asp since it is likely to have more accurate link metrics. See data for www.site.com instead?")

[ThompsonPaul]

Glad that was useful, Kimberly. Many times, folks just want a "when this happens, do that" kind of response, but usually in SEO there so many "it depends" aspects to an issue that it's vastly more useful to understand the Why, so you can work out the solution to the specific issue yourself. Sounds like you're that kind of person too.

That said - this discussion seems to be going round and round on you. If you want to PM me your site address, I can get a handle on your exact situation and try to explain exactly what's happening. I have my suspicions, but don't want to confuse the issue further with speculation.

Paul

[Cyrus-Shepard]

Hi Kimberly,

The rel="canonical" is a good option when you can't 301 a page. If you can 301 a page, such as /index.html or some other duplicate version, it's usually a better way to go. As you said, it removes all doubt about where to send visitors and link equity.

Both rel canonicals and 301s pass about the same amount of link equity (thought to be around 85%) so it's safe to use either.

Canonical tags are usually best when you have lots of parameters needed to render the page, like homepage?color=red&length=long&manu=apple. In this case it wouldn't make much sense to 301, and the rel canonical is easier to implement.

Hope this helps. Best of luck with your SEO.

[kimmiedawn]

Cyrus - Thanks. A familiar face - I've seen it attached to many an article that I've read;)

While I have you here, maybe you can answer another question related to the situation that has me slightly nervous:

I've read warnings about creating a loop with a 301 redirect - and I keep being pointed to default.asp in Open Site Explorer (with the message that the url I entered - www.site.com- redirects to /default.asp) but any redirect checkers I've used don't show that. Would it be obvious if I didn't implement it correctly? It looks fine in the browser.

Would a rel canonical tag in addition to a 301 redirect be a good idea?

[kimmiedawn]

In case this is helpful to anyone else reading this post, here is the code I am now using in the .htaccess, which seems to have eradicated the double redirect (thanks to help from phranque and lucy24 at webmasterworld):

RewriteEngine On

RewriteCond %{THE_REQUEST} ^[A-Z]{3,9}\ /([^/.]+/)*(index.html|default.asp)\ HTTP/

RewriteRule ^(([^/.]+/)*)(index|default) http://www.site.com/$1 [R=301,L]

RewriteCond %{HTTP_HOST} !^(www.site.com)?$ [NC]

RewriteRule (.*) http://www.site.com/$1 [R=301,L]

[Cyrus-Shepard]

Looks like you found the solution below. Nice work!