Just a quick tip tonight, since technically I'm supposed to be on vacation. One great way to help improve some of your rankings on a blog, content site or virtually any domain with lots of pages on similar topics is to eliminate the keyword cannibalization by 301'ing multiple pages relevant to a query/keyword to the highest ranking page. Here's a quick example:
For the query - SEO Pricing - SEOmoz has two results in the top ten:
If we were to redirect one page to the other, we'd have a considerable amount of additional ranking power, and probably move up the results considerably.
This tactic applies well to blogs where old posts might be less relevant than current ones. It also works on e-commerce sites where old products or product categories can be brought together. To find these potential joiners, just run a search like seo pricing site:seomoz.org. You can see three obvious candidates to be combined into one. We're not going to do this because each of these has individual value, but maybe in 6-12 months, I might re-direct the oldest of the bunch as its relevance will probably be limited.
Another quick example:
This query - Ashland Oregon Wineries - shows that AshlandWineTours.com has several pages they could conceptually combine (as they appear to be targeting the same keywords on all those pages). They could even 301 those pages, then re-create new URLs for more targeted content if they so desired.
Of course, this can be a terrible idea from a user perspective, so be careful how you implement it.
p.s. Mystery Guest and I saw the Simpsons movie tonight - tons of fun...
Now this is something I hadnt previously thought upon... brilliant! It makes perfect sense too... that way if more and more people did this, relevancy would certainly increase alongside currency as the older files would be 301'd to the newer relevant stuff.
Very interesting, and certainly something I'll be adding into the pot when I see a client where this could be done!
Thanks Rand!
It's a new idea for me too, but I'm not sure it's something I'd do. The page being redirected would have to be one I saw very little use for. Besides, I love seeing those multiple listings. They're rare enough that I don't think my response to getting a pair would be to think about replacing it with a single.
I wonder if this would work: you set up a 301 from the older URL, but you take the content from that page (some or all, whichever works better) and add it to the other. Obviously, if one page is completely out of date, it doesn't deserve to be kept, but if it's relevant, it might actually add to the relevance of the other.
Another thing to consider is that the reason that second page is ranking well may have little to do with links. If that's the case, and it's all about the content on the page, then 301ing from it to the other page will have a much smaller effect.
I think this is a great point - you need to analyse WHY that second page is ranking. Once you've determined that - then you can make a decision on whether to 301 or not.
Great advice Rand. We tell clients to do this all the time. There is a very close relationship between duplicate content and keyword cannibilization
In both cases, you are wasting link juice on pages that will never be highly ranked in the search engines. Even in the case of keyword cannibilization, the best you could expect is an indented second result, andI would much rather move 1 listing several places up in the rankings than have 2 listings further down.
So here is a question for you - If you have a client that absolutely will not eliminate a duplicate article, do you nofollow links to it, noindex it, put it in robots.txt, or some combination of these?
Our standard advice is to combine an entry in robots.txt with a nofollow on all links to it. The idea is to save the "crawl budget" and the link juice at the same time.
This is a very good idea and something that is simple yet not very obvious to novices like me
The data that I have seen on ranking CTRs suggest that if you have the #4 and #5 slot, even just getting a single listing in the #3 slot would get you more traffic.
More generally, of course there are times to do this, and times when you don't. A lot of the times we end up doing this when we find clients that have relatively low value pages that are competing for the same keywords as higher value pages.
Sometimes these are index pages, or pages with tidbits of additional information. Net-net, when you find such keyword cannibilization on low value pages, and we find them all the time on our client's sites, it is absolutely worth considering their integration/removal/redirection to the higher value page, or, if you must leave it in place for the user experience, NoFollow your links to it, and put it in Robots.txt.
What do you do with the content of the old page that is now being redirected to the new page?
You definitely don't want to lose the content and potential long tail matches that the old post got. What is the most effective way to deal with the content?
As I noted above - you can either combine it with the content on the page you're re-directing to, or, if you don't see it as valuable and relevant any longer, simply discard it. You can look at your stats to determine how much value (if any) it has from a "long tail" perspective and whether 2-3 visitors a day on long tail searches is more valuable than the potentially higher rankings for the high traffic terms.
Trying to concentrate on the issue from a content perpective rather than linking perspective. should'nt we check whether the old page reanked well before the newer page with new content was added? If it did, it might be related to links.
If it didn't (as was the case with my test), then it might be content (ofcourse bare minimum pagerank need for appearing and regular crawling) that caused the uplift rather than the links.
that is a great tactic if the situation calls for it. so many times it is abou t the situation at hand. play the cards that you are dealt. with some clients "more" listings are "better" listings.
Interesting.
How much more weight does the 301 redirect give versus a link from the old page to the new page?
If 301 is considered permanent by search engines, it would be all of it, correct?
Hi Cody - Assuming I understand you correctly, linking from the old page to the new page will pass some link juice to the new page, but a 301 redirect should pass all of the link juice to the new page.
Since more than a few folks got confused about this, let me just point out that you don't need to "sacrifice" the content of the re-directing page - that's kind of crazy. Instead, you could just add the content of the old page (or less relevant page) to the higher ranking page, so users would still see all of the benefit.
Also - with the broken-out link of two top listings - I'd say you need to make a judgement call on it. If you're ranking #15, #45 and #191, moving one of those to page 1 is far better to me than hoping you'll get the broken-out listing. This would, to me, even hold true if you had spots 5 and 6 for a very competitive SERP - earning that #1 spot could be done several ways, but unless you were confident about your ability to achieve it quickly, I'd shoot for re-directing one, moving content over logically and pushing to be #1 with one listing, then worrying about trying for a second, broken-out result.
So in your example above, you want to combine the content of the second page to the first page, moving your #4 and #5 slot to a single #1 slot? Each post presents the information in a different way, and there is additional value from all of the comments in each. I don't see how you could not sacrifice some of the content from the second post, otherwise it would be a jumbled up mess. Also, you could be ranking for a different set of queries on the second post than you are on the first, possibly eliminating some of those if you moved it.
Yep - as with many things in the SEO world, you'd have to make a judgement call. If we looked through our stats and saw that lots of other search referrals had sent individual traffic to that page, re-directing it might not be a great idea.
Good points to bring up as potential downsides to the strategy.
I would define this as a bad idea. You might be gaining a few spots for the fist page listed, but losing all exposure for a lot of alternate queries that your second page may have been ranking for while the first page was not. Those two pages are quite a bit different and provide different examples of pricing, each with useful information. Why completely remove the value from one?
Depends, in this particular case you might be right, but consider this: you have a page with some data on a particular field of your choice.
This data is valid for the period 2005-2006. Now you make a second page a year later, with data about the same field, which is valid for 2005-2007.
You could know upfront now that you would need to make sure to use all the keywords that this other post ranks for, and you could 301 the old post in to the new one... Makes sense, doesn't it?
Sort of. If you knew upfront it would make more sense to me to update the old page with new information. However if it was unintentional and after the fact then moving it would make sense. However, I see using this as a rare occurrence and should only be done to fix a mistake.
I don't agree, of course you can do it on purpose, especially on a blog where usually, updates will be new articles...
I suppose it could go either way on blogs. Sometimes I simply update the post, other times it warrants a new post. You would need to analyze each case. Sometimes I would want double listings; other times in more competitive phrases I would want a boost in rank for that particular post topic.
Great idea but doesn't having the multiple listings in Google like this increase the likelihood that someone will click on your link?
So if that's true (which it mightn't be) this technique could be removing something very useful gifted to you by Google.
Phil: the CTR of a #1, #2 or #3 ranking is way way higher than that of a #7 and a #8 ranking combined :)
A nice way of doing this on WordPress blogs is using Objection Redirection.
This guy has a nice attitude Joost :)
Hope his plugin has a more positive contribution
Hehe, I think you missed some cynicism there, as Dax, the writer of it, works for Greg Boser :P
Is true that the CTR is lower at #7 but way no just try to work hard to get better positions on Google, what about the user, what about your content (old post) maybe you receive visits with other keywords to that post.
That could be good only if the user is not important to you.
Hey don't do it for everything, just for the posts you think you can miss :) You know that from your analytics don't you? :)
I agree. If you could move up to the top 3 spaces without redirecting you would have a very high CTR rate because you would still have two listings. But Rand's method could be the best thing to do in many circumstances.
Rand & Others - what is your best guesstimate as to how long after implementing the 301 that you'll be able to notice a rankings boost?
In my experience, 301 redirects for an entire domain can often take quite a while to really "kick in" and typically doesn't go quite as smoothly as it could/should in theory. I don't believe I've experimented much though with just one or a few single pages redirecting to another page on the same domain.
Are we talking days, weeks, months??? Obviously "it depends" but I'm going for a general feel here if you will :)
Ephricon - My experience is that Google usually absorbs and integrates a 301 in a few weeks time, but of course there are many factors that drive this.
On a related note, even when we redirect an entire domain, we have seen response times of a few weeks. I would be really interested in hearing more about your experiences where it did not happen that quickly for you.
Eric: "a few weeks" time depends entirely on the domain doing the redirecting... If that's a PR6 or 7, it might only have to take 4-5 days, if that's a PR2 it might take way longer...
Rand, it would be great if we could test the theory by acutally doing the 301 on the two pages ranking for SEO pricing above!
Another option is to add additional links between the articles (if links don't already exist), as it is obvious that google already interprets them as related content.
That way you don't have to sacrifice the contents of one of the articles, can could still improve both their ranks. :)