Way back in October 2002, Danny Sullivan wrote:
Now I can make my advice about the meta keywords tag even easier. Just don't use the tag at all! Obviously, if you personally find it or believe it to be useful, keep doing so. But I suspect it's just a waste of time, for most people.
But anyway, lets see what $200 of Search Engine Optimization buys us these days. Here are the Meta tags added by a SEO company to a friend's website:
<meta name="key Phrase" content="[...snip...]">
I'm not sure what "key Phrase" is supposed to do, it's not part of DC Metadata and a quick Google doesn't bring anything up. Not looking to good so far.
<META HTTP-EQUIV="expires" content="0">
Ah ha! A Expires HTTP-EQUIV tag, this might be useful. According to RFC 2616, the Expires header should in RFC 1123 format so this isn't too helpful. I'm guessing that they mean that it expires RIGHT NOW.
<META HTTP-EQUIV="pragma" content="no-cache">
Looks like they really don't want us to cache the page. This is a shame because it's not likely to be updated often.
<META NAME="Classification" CONTENT="[...snip...]">
Equally as useful as "key Phrase". It actually contained the same text, but with "information" appended. Probably trying to target people who like to type in extra words into Google.
<META NAME="Rating" CONTENT="General">
It's a business to business site. Safe for kids looking for industrial products.
<META NAME="Revisit-after" CONTENT="28 days">
So we don't want you to cache the page because it's updated really frequently, but search engines shouldn't come back in the next 28 days. Hmmm.
<META NAME="Updated" CONTENT="28 days">
I guess they really mean it.
<META NAME="Distribution" CONTENT="Global">
It's not called the World Wide Web for nothing!
<META NAME="ObjectType" CONTENT="Document">
I'm not sure why we need to point this out.
<META HTTP-EQUIV="Content-Type" content="text/html">
Not totally useless I guess. Probably would be more useful if you told me the character set.
<META NAME="Robots" CONTENT="ALL">
We don't want to turn away any robots accidently. This would be the default, right?
<META NAME="Publisher" CONTENT="[...snip...]">
In case the domain name didn't give it away.
<META NAME="AUTHOR" CONTENT="Text/notepad (TKMSA;JBH1) [Netscape, MSIE,AOL, Mosaic]">
I guess they are proud notepad users?
<META NAME="Search Engines" CONTENT="Google, AltaVista, AOLNet, Infoseek, Excite, Hotbot, Lycos, Magellan, LookSmart, CNET, yahoo">
We'll let all robots come visit but we only care about these search engines. Maybe we think that if we specify the search engine they'll rank us higher?
<meta http-equiv="Content-Language" content="EN">
A real tag! I think it should be "en" though.
<META name="MSSmartTagsPreventParsing" content="TRUE">
Stop those evil Microsoft products adding "smart tags". But wait, do any Microsoft products support "smart tags"?
<meta name="copyright" content="[...snip...].com 2004">
Might as well have this in the meta tags and on the page. Can't hurt.
<META http-equiv=Content-Type content="text/html; charset=windows-1252">
We REALLY want you to treat this as text/html. Let's specify a character set this time.
Most SEO companies are pond scum. This is an example of one fleecing somebody out of $200 to "improve the priority of the website on search engines". Be warned.