Google & Yahoo Treat Undelayed Meta Refreshs As 301 Redirects

Undelayed Meta Refresh Gets Treated As Permanently Moved 301 Redirect

Author: Jonas Jacek03/ 09/ 2007

Search Engine Optimization

At times when you use a meta refresh it can often trick the users of the site to see pop ups by actually getting onto pages of what they call low life free hosting. The search engines do not see every single instance of meta refresh as spam. When webmasters move their free hosted sites to their own domains they usually depend upon a redirect to a new location.

Yahoo says that they treat zero meta refreshes which is a redirect with a change delay in zero seconds. They use the formula of

META Refresh: <meta http-equiv=”refresh” content=…>

is recognized as a 301 if it specifies little or no delay or as a 302 if it specifies noticeable delay.

While Google is working on rewriting their documentation, they are using a version of their own help docs in the Meta refresh which is not yet put out for all to see. The Google Mini sees all meta refreshes as 302 redirects. Google uses a META tag that specifies http-equiv=”refresh” is handled as a 302 redirect. But that’s handled differently on the Web. The best idea is to use 301/302 redirects directly whenever possible. After that the next best redirect method is to do a metarefresh with 0 seconds delay and therefore is a for a 301 permanent redirect.

If you are going to use the 301 redirect try not to change the Meta refresh only use this method when you really have no other choice. This next function is used when there is a no delay in meta refresh which is sometimes called a poor man’ 301 redirect.

<html>
<head>
<title>Moved to new URL: http://example.com/newurl</title>
<meta http-equiv=refresh content="0; url=http://example.com/newurl" />
<meta name="robots" content="noindex,follow" />
</head>
<body>
<h1>This page has been moved to http://example.com/newurl</h1>
<p>If your browser doesn't redirect you to the new location please <a href="http://example.com/newurl"><b>click here</b></a></p>
</body>
</html>

This will work as long as the server will handle and show the content under the old URL. The bot from the search engine will crawl the stats and should not be listed in the URL under any 404 errors. If these do show under the Not Found header then something horrible has happened. Most likely it will occur on the free hosting end. If you control the account, When you have control over the account, you cannot delete the page because the search engines will want to see it whether it is a redirect or not.

Excursus: When a search engine crawler fetches this page, the server returns a 200-OK because, well, it’s there. Acting as a 301/302 does not make it a standard redirect. That sounds confusing to some people, so here is the technical explanation. Server sided response codes like 200, 302, 301, 404 or 410 are sent by the Web server to the user agent in the HTTP header before the server delivers any page content to the user agent (Web browser, search engine crawler, …). The meta refresh OTOH is a client sided directive telling the user agent to disregard the page’s content and to fetch the given (new) URL to render it instead of the initially requested URL. The browser parses the redirect directive out of the file which was received with a HTTP response code 200 (OK). That’s why you don’t get a 302 or 301 when you use a server header checker.

When a search engine sends a bot or a spider to crawl or retrieve the pages it is just the beginning of a very complicated process. The search engines are on a larger scale system that will makes use communications from millions of speculated programs. The spider will have zero to do with any indexing but it may follow the server side of the redirects but if you are using Meta refreshes then that may not happen.

Removing a redirect page is done by a process like vertical indexes pulling out there fodder. This can mean that the old pages of the site will get deindexed before the newer URL is used in the search indexes. If you modify anything during this process then you run the risk of getting deindexed.

So keep all redirect permanent changing them will make the indexing process harder.
The zero Meta refresh will work as a 301 redirect due to the fact that the search engine will be treated as a permanent redirect. But not in 301 form.
There has been so much abuse from spammers that the 301 may be seen as less reliable on the server side and will be sent into the HTTP header. There are several ways in which you can do that. Sebastian from Sebastians-Pamphlets.com collected a few:

The page title says that the resource was moved and tells the new location. Words like “moved” and “new URL” without surrounding gimmicks clear the message.
The zero (second) delay parameter shows that you don’t deliver visible content to (most) human visitors but switch their user agent right to the new URL.
The “noindex” robots Meta tag telling the engines not to index the actual page’s contents is a signal that you don’t cheat. The “follow” value (referring to links in BODY) is just a fallback mechanism to ensure that engines having troubles to understand the redirect at least follow and index the “click here” link.
The lack of indexable content and keywords makes clear that you don’t try to achieve SE rankings for anything except the new URL.
The H1 heading repeating the title tag’s content on the page, visible for users surfing with Meta refresh = off, accelerates the message and helps the engines to figure out the seriousness of your intent.
The same goes for the text message with a clear call for action underlined with the URL introduced by other elements.

Other ways to make it easy for a search engine to understand meta refresh redirects correctly are:

The page title says that the resource was moved and tells the new location. Words like “moved” and “new URL” without surrounding gimmicks clear the message.
The zero (second) delay parameter shows that you don’t deliver visible content to (most) human visitors but switch their user agent right to the new URL.
The “noindex” robots meta tag telling the engines not to index the actual page’s contents is a signal that you don’t cheat. The “follow” value (referring to links in BODY) is just a fallback mechanismn to ensure that engines having troubles to understand the redirect at least follow and index the “click here” link.
The lack of indexable content and keywords makes clear that you don’t try to achieve search engine rankings for anything except the new URL.
The H1-Tag (heading) repeating the title tag’s content on the page, visible for users surfing with meta refresh = off, accelerates the message and helps the engines to figure out the seriousness of your intent.
The same goes for the text message with a clear call for action underlined with the URL introduced by other elements.

Also interesting to know: Yahoo! Search treats meta refreshes of less than a second as a 302 redirect. This means Yahoo! found the page, but eslewhere. Meta refreshes of greater than a second are treated as permanent (301) redirect. Maybe a Yahoo! Search Bug. The other way round would be more understandable (see up).