301 redirects: not so good for SEO after all?

When you’re changing the structure of your website, changing your primary domain or creating better URLs for SEO purposes, you want to keep the values of your old URLs and move them to the new URLs. Luckily there’s an easy solution for that: 301 redirects. 301 redirects let your server tell the visitor (or search engine) that the page they are trying to visit has been moved permanently to another location. Visitors will be sent to the new location directly. Search engines will try to remove the old URL out of their index and move the value of that page to the new location. For some time now Google is telling us that 301 redirects are the proper way of redirecting your pages. However 301 redirects seem not so good for SEO after all.

301 redirects and PageRank

An import measure of the value of a page for Google is the PageRank. When you’re moving your pages you would want all of the PageRank from the original pages to be relocated to the new locations you specify in your 301 redirects. But in an interview in January with Eric Enge, Matt Cutts confirmed that with using a 301 redirect there might be some loss of PageRank. Too bad Matt didn’t really explain why. In my opinion moving a page, shouldn’t make that page of any less value. It’s still the same page with the same information people originally linked to. Imagine changing your company name or optimizing you URLs for SEO purposes. You would be forced to change all the URLs of your website and you would lose a little bit of PageRank for every redirect. But that means losing a little bit of PageRank for every page on your website and decreasing the authority of your whole site.

301 redirects and relevancy

Besides authority in the form of PageRank links also pass relevance to the page they’re pointing to. One way links pass keyword relevance is through the anchor texts of the links. With a redirect there could be some problems however, since the link doesn’t point directly to the new page but through a redirect. Google says in a video that anchor texts typically do flow through 301 redirects. However they don’t promise this will always happen based on the trust they have in specific redirects. They’re giving a signal here there are situations where anchor texts won’t flow through a redirect.

In a test Dave Naylor for example proved that in a specific case an anchor text wasn´t passed through a 301 redirect. Over a year ago Patrick Altoft noticed the same behavior for some 301 redirects. Google could do this to prevent Google bombing or maybe to decrease te value of redirecting links for link building purposes. The consequences of not passing anchor text relevancy through 301 redirects however can be enormous. When changing your domain you could lose all relevancy you carefully built with the use of anchor texts. So search engines should be really careful with this.

Practical examples

In a practical case Mark Lavoritano proved that moving domains had no significant impact on the search traffic Google generated. However the search traffic from Yahoo! And Bing decreased dramatically. So while Google is telling us 301 redirects could not pass all the value and could cause a decrease in findability they seem to handle it the right way. They´re probably just trying to tell us not to buy links and redirect them to our websites to increase our findability.

What to do when changing your URLs?

When you need to change you URLs there is no real alternative for using a 301 redirect. So 301s are still the best way to redirect but consider the fact that you could lose authority and relevancy along the way. Where Google seems to support the function of 301 redirects very well, other search engines aren’t that good. This forces you to consider the need of changing your URLs and makes you think about creating URLs that you don’t have to change any time soon. W3 advises to create good URLs by leaving out:

  • Authors name– authorship can change with new versions. People quit organizations and hand things on.
  • Subject. This is tricky. It always looks good at the time but changes surprisingly fast. I discuss this more below.
  • Status– directories like “old” and “draft” and so on, not to mention “latest” and “cool” appear all over file systems. Documents change status – or there would be no point in producing drafts. The latest version of a document needs a persistent identifier whatever its status is. Keep the status out of the name.
  • Access. At W3C we divide the site into “Team access”, “Member access” and “Public access”. It sounds good, but of course documents start off as team ideas, are discussed with members, and then go public. A shame indeed if every time some document is opened to wider discussion all the old links to it fail! We are switching to a simple date code now.
  • File name extension. This is a very common one. “cgi”, even “.html” is something which will change. You may not be using HTML for that page in 20 years time, but you might want today’s links to it to still be valid. The canonical way of making links to the W3C site doesn’t use the extension.
  • Software mechanisms. Look for “cgi”, “exec” and other give-away “look what software we are using” bits in URIs. Anyone want to commit to using perl cgi scripts all their lives? Nope? Cut out the .pl. Read the server manual on how to do it.
  • Disk name – gimme a break! But I’ve seen it.

You can use creation dates however, because they won’t change afterwards.

About Jeroen van Eck

Jeroen van Eck is a consultant search engine marketing at the online marketing company E-Focus in the Netherlands.

23 thoughts on “301 redirects: not so good for SEO after all?

  1. A 301 redirect is an extra level of indirection. I dare say it acts like a link, a link to a page that contains only one other link to the destination page you are redirecting to; so although all value will be transfered to that one link, it is also subject to the ‘random surfer’ effect, probably meaning approximately 15% loss in value if you go by what is in the original PageRank paper.

    I think you are correct though, and that redirects should not be taken lightly as a quick solution; the exact risks and benefits have to be weighed up before action is taken.

  2. I’m with you here. Why on Earth should moving a page lead to any loss of authority? To use a real-world example, if we move our offices to the vacant spaces two floors up, would our brand lose credibility?

    It’s ludicrous really.

  3. Anchor text won’t flow through a 301 redirect when you do something like redirect all URLs in an entire folder or an entire site to a SINGLE new page URL.

    It also doesn’t flow when the content at the new URL has no relation to the content previously found at the old, now redirected, page URL.

    There are a few other cases too. And, I think that’s the correct way to handle those things.

  4. The other concern with 301s is sometimes how long they take to “work”. A 301 seems to work by two seperate processes that SHOULD work seemlessly but often do not. The first process recognises that the old page is no longer at the url it was at, in my experience the page immeadiately drops out fo the index at this point. A second process, some time later, comes along to index the new url.

    There is sometime a huge delay between these two processes; I’ve personally seen a delay of up to 5 weeks! Now what causes the delay in these two joined processes I haven’t quite figured out yet, perhaps someone else can comment on that.

    I’ve also heard anecdotal evidence of 302 redirects working much better than 301s (which goes completely against what we all think we know) but this is not something I have had any experience of yet.

  5. @g1smd What are you basing that on? A redirect is shown to a robot when they’re requesting a URL. How would they know if it’s based on a redirect for a complete folder? They just get the new location.

    @toot I’ve seen cases where the old URLs keep existing in the index for a few weeks even when the new URLs are already in the index. But that’s the opposite of what you’re saying. In my experience this happens more often (in Google) than the other way around. I don’t see how a 302 redirect could be working better than a 301 though. Do you know anything more about this?

  6. Why would you have this title that 301 is not good for SEO, when after all it is better than losing your status altogether….???? It is better to have it than NOT to have it …

    A pointless article, really…. 🙁

  7. @Helen It’s true that it’s better to have it than not to have it. But what I want to highlight is that redirecting your URLs still has a negative effect on SEO. So you should reconsider carefully whether it’s necessary and you have to think about how your URLs should be shaped to avoid redirecting in the future.

  8. I hope my comment is not taken out of context.

    When I’m looking for SEO advice, I can only rely on the most trusted advice.

    The fact that your articles are full of spelling and grammar errors makes it harder to trust your advice; thus, if you want the respect of the audience you hope to attract and keep, editing posts is very highly recommended.

    Just trying to be helpful


  9. Hi Kevin, please bare in mind that his native language is not English. We’ll be looking at the grammar a bit more, you get my word on that, but you can also trust that the things he is saying are not rubbish :).

    Thanks for the tip, we’ll look into the grammar bit

  10. @Kevin Spelling and grammar has nothing to do with SEO. SEO doesn’t require that.

    @Jeroen van Eck – yes 301 can have negative effect on SEO but that is still the best redirect we have in place. In terms of other search engines not being able to read it (who cares) Google is the biggest traffic generator (currently expect for China and Russia – if I have missed another country then apology). So in all honesty I wouldn’t put curtains on 301s. From my experience I have seen positive things happen because of this BEAutiful 301 redirects :).

  11. 301 redirects are the ONLY efficient way to resolvea URL change in SEO. I’m glad no one mentioned canonicals which:
    – Are hints and not directives which means that Google can ignore them
    – Can screw up rankings if badly implemented eg. if because if a typo they resolve to 404 pages.

    I wouldn’t worry about losing revelance as one or two links that don’t pass relevance cannot be conclusive. Dave Naylor just tried with a few anchor text values, usually an established site would have large volumes of anchor text values and in my experience these never get lost. 301s are a no brainer as there is still no alternative…

    As for Google and Yahoo, they need to get better if they want to earn a greater market share.

    @Jeroen van Eck – Good example of a good link bait article! If 301s aren’t so good for SEO why don’t you tell us what is it good then? Not changing the URLs at all?

  12. @Saurav @Modi I agree with both of you that 301 redirects are best way to handle moved pages. However with this post I want to warn everyone that it shouldn’t be used too freely because it does have negative impacts on your SEO.

    As Modi says the better alternative is not changing URLs at all. Therefore when creating or changing URLs think about how to build the URLs so they don’t need to be changed afterwards. Follow the instructions given by the W3 mentioned in this article.

  13. If you want to see how rel=”canonical” handling is worse than using a proper 301 redirect, look no further than the indexing of pages which have been moved around within any MediaWiki-based wiki.

    Accessing a page that has been moved merely shows “redirected from…” text and a link at the top of the content page and a rel=”canonical” tag in the head section, but crucially the content continues to be served at the old URL as well as being served at the new URL.

    It’s even more fun when Google picks the non-canonical URLs and uses them for your sitelinks. Even after blocking the old URLs in the WMT sitelinks option and installing a hard 301 redirect in .htaccess, the sitelinks remain stuck showing the old URLs.

  14. Another thing to consider is using Google Webmaster Tools and check for the errors Google is getting with 404 pages. This can be useful for a website not even changing any URL’s because 404 problems sometimes just occur. If you are changing your website this will tell you what 404 pages Google got and the date that they occurred. Fixing these will help Google and the people who visit your site.

  15. I think this is a useful article as it raises possible problems with 301 redirects. Sure there is no better solution but it is better to discuss the issues at hand.

    I would like to know what everyone thinks about links to a redirected page. If you have built up a lots of links to inner pages and then you 301 those pages, should you leave the external links pointing to the old URL’s or the new pages?

  16. @Jeroen Nice post, I had one query though, in my website, I can classify the 301 redirects in to two types:
    1. Internal 301 redirects (internal links “A” on the same domain getting redirected to link “B” on the same domain.
    2. External links (links from other domains) linking to the link “A”

    I can’t control the external links but I can surely control the internal links, therefore I plan to replace all the 301 redirected URL’s on my website with the actual URL’s, hence taking care of the internal 301 redirects and for the external 301 redirects, I plan to maintain the 301 redirects in the .htaccess file, when I run the “Cache” command on Google for the link “A”, it shows me the redirected URL “B” cached.

    Given the above scenario, I wanted to understand if my decision regarding replacing the internal URL’s on my website will have any impact/side effects on the SEO for the website.

  17. I can’t see there really being any alternative to a 301 redirect? If it’s necessary, then it’s going to be unavoidable (e.g. if a company needs to change it’s domain/company name, it shouldn’t be punished for it).

  18. I have seen some do redirects and the PR tool bar went up immediately, other I have seen do it and nothing happens.  There is something strange going on with 301’s when it works from some and not others.  I am sure in the future Google will be able to prevent any passing of value from one site to another.

Comments are closed.