| 12:46 pm on Sep 3, 2008 (gmt 0)|
The quick answer is no - words in meta keywords are highly unlikely to have any noticeable effect on performance, negative or otherwise. They are largely ignored by Google. If they contained obscenities or similar then you might have issues but otherwise, I would just omit them as your time is better spent elsewhere.
As for duplicate content, the term is normally used to describe large amounts of duplicated text, either an entire page duplicated exactly at a different URL, or substantial amounts of text duplicated elsewhere. This is not an issue in meta keywords, although you should certainly avoid duplication within the title and meta description elements.
| 1:05 pm on Sep 3, 2008 (gmt 0)|
|They are largely ignored by Google |
A brief adjunct question about the word "ignored":
From Google's point of view, does the repeating of a specific word in meta keywords multiple times have any actual impact on a page's "keyword density"? and even if it does, does keyword density matter anymore?
(In other words, if that old meta tag is literally "ignored", then they would not be tallied in the keyword density percentage)
| 1:13 pm on Sep 3, 2008 (gmt 0)|
to me why have them repeated as it may cause issues so why even risk it. Stuffing the words in the keywords can be considered spam and really know one knows if Google actually ignores them or not so my answer is yes it could be considered spam so don't do it.
| 1:42 pm on Sep 3, 2008 (gmt 0)|
|The quick answer is no - words in meta keywords are highly unlikely to have any noticeable effect on performance, negative or otherwise. |
Maybe for Google but what about Yahoo!?
How should I write my keywords?
|Enter your keywords in order of importance, and be sure to use keywords that actually appear in your page content. Don't repeat keywords more than twice (and not consecutively!), separate your keywords with commas, and keep the list to under 256 characters. |
| 1:53 pm on Sep 3, 2008 (gmt 0)|
I've seen some pretty spectacular meta keyword elements over the years (5k meta keywords anyone?) but ever never seen any real impact on Google results in changing them, cleaning them omitting them, whatever.
IMO the complete lack of standardisation in terms of what should be included in meta keywords tags means that webmasters with clean sites include 'spammy' keyword elements in all innocence - believing that, for instance, all of the output from a keyword suggestion tool would be helpful, or the entire textual content of the page separated by commas. For that reason, I think it is a wise decision for a search engine to ignore or place little weight on the content of meta keywords.
| 2:14 pm on Sep 3, 2008 (gmt 0)|
|I've seen some pretty spectacular meta keyword elements over the years (5k meta keywords anyone?) |
Me too! I think we've probably seen some of the same ones. :)
|But ever never seen any real impact on Google results in changing them, cleaning them omitting them, whatever. |
"Any real impact". < Okay, that one covers you.
I've never seen any documented testing in this area other than the FAILed methods of using some obscure keyword in the keywords metadata only which isn't a test at all.
|For that reason, I think it is a wise decision for a search engine to ignore or place little weight on the content of meta keywords. |
"Or place little weight on..." < That one covers you too! :)
If I were a search engineer, I'm going to utilize that element if it is there and if it is relevant. I think my algo is smart enough to determine if that element is being used properly or improperly.
The keywords metadata element has been around since the dawn of SEO. So has the description metadata element. I remember a few years ago some hopped on the bandwagon that meta descriptions had little value. I think we all know what happened there in the past 12 months? To top that off, we have videos circulating on YouTube that state the meta description has little to no value in the overall scheme of things. < Can you believe that?
Here's how I look at it. I've been using that element since the beginning. I still use it to this day. As long as I keep finding references to it at the W3, Search Engines, other indexing services, I will continue to use it.
Can you tell me for sure that Google, Yahoo!, Live and Ask do not calculate the words within the keywords metadata element? Are we absolutely, positively sure?
Take the <title>, meta description, <h> and first content after the opening <body> element. Extract 3, 6, 9, 12 words and/or phrases. Assemble them in order of importance, separate them with commas and be done with it. Takes about a minute and once it is done, you typically don't fiddle with them.
Many of the sites I've worked on have a separate area for entering keywords. These are used for on site search functions and to populate the keywords metadata. It's a no brainer and from my perspective still good practice. You just never know...
| 2:32 pm on Sep 3, 2008 (gmt 0)|
|I think my algo is smart enough to determine if that element is being used properly or improperly. |
I like that attitude, and I like to think that if a person exceeded the number of allowed characters, the counting would simply stop. So, having 512 or 800 or 5k characters would make no difference -- it simply stopped at 256 and that's that.
Whether that's the case or not I do not know for sure, but in a perfect world, my "smart algo" should be able to figure that out.
I also like to think that while the meta keywords may not help one bit with ranking position, what it might still do is help in some minor way to confirm for the algo what the page is about (the "theme"). Thus, like you, I've been using it since 1996 without any expectation of getting a boost, but rather erring on the side of caution (and habit!).
| 2:36 pm on Sep 3, 2008 (gmt 0)|
Every few years I look at meta keywords again. Then I look at pages on my site put up back in 1996 with no meta tags at all and compare with recent pages put up with meta tags... and don't see much performance different ON MY SITE. YMMV. So far I haven't discovered conclusive proof these tags (keywords, description) make that much difference, so I wait a bit more and fiddle a bit here.
A recent test regarding a static page unchanged since 1999 was interesting. Removed all meta statements. All. Changed nothing else. Page jumped up. I have about 200 pages from that same period and am thinking of doing the same, just to see what happens.
| 2:45 pm on Sep 3, 2008 (gmt 0)|
|A recent test regarding a static page unchanged since 1999 was interesting. Removed all meta statements. All. Changed nothing else. Page jumped up. I have about 200 pages from that same period and am thinking of doing the same, just to see what happens. |
I just recently saw a similar comment concerning meta descriptions. It was claimed that after removing them, positions improved. My conclusion? The descriptions may not have been properly crafted and what the bot found after the opening <body> element had more weight?
When it comes to the keywords element, if I had to guess, the value of it lies way down there at the bottom of the list. But, when you go through the list and can check off everything that is there, you've covered most, if not all of your bases. That's the way I like to look at it. Leave no stones unturned as they say.
There are also quite a few other elements that are not discussed in depth around here that I use that appear to have an overall impact on things. Just little tidbits of advice the W3 give after you read a page that is a 1/4 mile long! ;)
| 5:29 pm on Sep 3, 2008 (gmt 0)|
|I've seen some pretty spectacular meta keyword elements over the years (5k meta keywords anyone?) but ever never seen any real impact on Google results in changing them, cleaning them omitting them, whatever. |
I thought it was widely accepted that Google ignored the Meta Keyword tag anyway?
| 6:47 pm on Sep 3, 2008 (gmt 0)|
|I thought it was widely accepted that Google ignored the Meta Keyword tag anyway? |
I think we should provide links to where Google states that in their Webmaster Guidelines so there isn't any confusion on the matter.
It was widely accepted a couple of years ago that meta descriptions had little value too. I challenge anyone to remove their meta descriptions today. :)
Now, I know this is the Google Search News Forum, but what about all those other search engines? You know, the ones we previously utilized? Why does Yahoo! have an entire page devoted to the development of the keywords meta element?
|From the Wiki: "In April 2007 that the relevance of having your keywords in the meta-attribute keywords is little to none." |
Notice the "little to none"? I've yet to find someone who has documented evidence to support this whole keywords thing. I know, the topic really isn't worth the time and effort. But, I want to know why the information that is out there that is recent still conflicts with what many are claiming. For one, Yahoo!'s focus on the element in their Small Business documentation.
Also, the tests I've seen to date are flawed, the ones where some obscure keyword is targeted in the keywords metadata element only. What kind of test is that? FAIL!
| 6:52 pm on Sep 3, 2008 (gmt 0)|
|It was widely accepted a couple of years ago that meta descriptions had little value too. |
That was before the snippet started to use the meta description so widely - in those days we had a description on the search results pages that looked like a ransom note.
That's the big difference between a meta keywords element and a meta description. The keyword tag is completely invisible, and the description gets surfaced on search results much of the time.
| 7:37 pm on Sep 3, 2008 (gmt 0)|
|It was widely accepted a couple of years ago that meta descriptions had little value too. |
Here are a couple of references from Google sources:
Read the full post + the comments below for reference to the keyword tag.
"Next we have two name values: keywords, which these days is mostly useless, ironically, and description, which is still somewhat useful."
As for the other search engines, the only other one I somewhat care about is Yahoo, and my sites do very well in their results without there being a meta-keyword tag on my pages.
If you spend a few minutes searching your favorite SE, you can find dozens of references to this topic
| 8:03 pm on Sep 3, 2008 (gmt 0)|
Thanks for those links. I noticed that Google's John Mueller said...
|You're right in that we generally ignore the contents of the "keywords" meta tag. |
The wiggle room word in his reply is "generally" and that pretty much lines up with what pageoneresults has been saying. If you get over the top spammy in the meta keywords element, I think that might trigger some kind of further look at what you're doing in other areas.
| 10:35 am on Sep 4, 2008 (gmt 0)|
well thanks for all the reply guys
| 1:37 pm on Sep 4, 2008 (gmt 0)|
I make it routine to fill in the meta keyword tags because Yahoo uses them and I can rank for synonyms and phrases that I do not wish to include in the text content of my page.
| 10:01 pm on Sep 4, 2008 (gmt 0)|
|...because Yahoo uses them and I can rank for synonyms and phrases that I do not wish to include in the text content of my page. |
The last time I heard it described by Yahoo, Yahoo uses them as a last option... ie, after it's looked at all other web content and found nothing, it will then reference words that appear in meta keywords tags.
Otherwise, the meta keyword tag, IMO, is tooth fairy stuff.
| 4:00 pm on Sep 5, 2008 (gmt 0)|
No one has mentioned, that g webmastertools complains about faulty meta tags with reportung :
duplicated meta tags - title
meta tags too short - descr
| 4:10 pm on Sep 5, 2008 (gmt 0)|
|Otherwise, the meta keyword tag, IMO, is tooth fairy stuff. |
Heh! She left a "C Note" under my pillow this morning. :)
| 7:14 am on Sep 7, 2008 (gmt 0)|
|She left a "C Note" under my pillow this morning. :) |
Did you maybe have some sort of side deal with her? ;)
This can't be about meta keywords. Her magic powers just aren't that strong.
| 5:51 pm on Sep 7, 2008 (gmt 0)|
|Did you maybe have some sort of side deal with her? |
I did, she paid me to provide supporting documentation for the keywords element in this topic. It really is all over the place. I use it by default and not based on Yahoo!'s perversion of it but my own which is to only use the primary keywords for the page, the top 2-5 and that is it. It's not something you invest a lot of time in and its usually the last thing to get populated after the meta description so it really is no big deal.
|However, no major search engine today claims to read the keywords which has raised the question of whether they are still needed at all. |
The above from the author at the Wiki on this subject. It pretty much sums up how I feel about it. But again, until Google, Yahoo!, Live, etc. put it in writing that the meta keywords element is no longer utilized in their algos, then the question will continue to remain, won't it?
200+ factors to determine the relevance of a document. What exactly are all of those factors? And, if you utilize all of those that are available to you, is it possible you can refine the relevance of your document just a wee bit more than another? Maybe I'm just too old school. It really is a habit. For the 10 seconds of my time that it takes to populate one manually, it just happens. :)
Hey, if I want to wear Bell Bottoms while waiting for the Tooth Fairy to drop another C Note under me pillow, that is my right, yes?
And by golly, did you know the thing can be used for other functions besides the SEs?
Ya, I'll agree, the message is loud and clear. This particular element is of "little value". ;)
| 12:11 pm on Sep 8, 2008 (gmt 0)|
|I'll agree, the message is loud and clear. This particular element is of "little value". |
And the keyword there is "little" -- so as you say, when the major engines officially say that the meta kw tag has "NO value", then I'll skip it. Until then, even "little" is worth my 10 seconds!
<Edit: fixed typo>
[edited by: Reno at 12:13 pm (utc) on Sep. 8, 2008]
| 1:17 pm on Sep 8, 2008 (gmt 0)|
Well pageoneresults in your own words
I would assume there are many of the 200 factors that add "little vale" but leave enough of the out and that reduces the "little value" effect that add up to either showing up or not.
|200+ factors to determine the relevance of a document |
I add them by default "little value' or not I add them to all my pages.
I would love to see a list of 200+ factoring to determine the relevance of a document as I can come up with a few but 200....
| 12:40 am on Sep 9, 2008 (gmt 0)|
It's now a fact of SEO life that 'location, location, location' matters.
Especially with Google, and, I suspect, the others will eventually catch on.
But using the keywords tag is a limited approach, with no evidence to support it.
For Google, at least, I've yet to see any evidence at all that the KEYWORD tag has any value unless it reflects the page content - and I'm not totally convinced, even then.
I'm beginning to suspect that for US sites, geolocation tags may make a difference - but the evidence is not yet totally convincing.
You need to think about your content above all; including zip/postal codes and telephone numbers.
(no change there, then)
| 2:46 am on Sep 9, 2008 (gmt 0)|
Back on the third I mentioned an experiment of removing meta statements from an elderly page unchanged since 1999, and a subsequent increase in ranking. I picked 10 pages from the 200 from that same time period and repeated the action: removing all meta tags. Five days later each page bumped up significantly. The keywords were appropriate and the description was appropriate for each page, nothing dodgy or over the top "widget, blue, round, company" and "Best blue widgets in round configuration from ZYX!"
I'll wait a week longer and see if the bump holds. If it does I might consider removing all meta statements for keywords and description site wide. The remaining meta statement choices seem to have some value for site management, so further thought is required.
| 12:44 pm on Sep 9, 2008 (gmt 0)|
|Back on the third I mentioned an experiment of removing meta statements from an elderly page unchanged since 1999, and a subsequent increase in ranking. |
I remember someone else stating that same thing about meta descriptions a couple of months ago. They went through and removed their meta descriptions. They stated that everything improved afterwards. Logicial deduction? The metas were not up to snuff and may have been causing more harm than good, referring to meta descriptions that is.
One or two people removing metas and seeing an increase in ranking doesn't do much for me. My first thought would be that they were not sufficient in their content and/or construction. I'd have to see before and after before believing anything related to meta description removal having an improvement.
| 1:26 pm on Sep 9, 2008 (gmt 0)|
I don't understand this sudden focus on meta-tags popping up everywhere. Only because the webmaster tools give off a small warning when your description tags are duplicate? Nowadays, the only benefit you will have from a good description-tag is improved CTR in the SERPs.
As for the keyword tag... It's a relic of the past. I think it's only useful for auto-populating the fields of some directory submission forms :)
[edited by: johnnie at 1:28 pm (utc) on Sep. 9, 2008]
| 1:28 pm on Sep 9, 2008 (gmt 0)|
I've got to agree with you, pageonresults, since I have so many cases behind be where ADDING good meta descriptions improved ranking and traffic.
However, I've never seen that happen with meta keywords in maybe 7-8 years. Of course, I haven't even bother with the experiment in quite a few years, either.
| 1:40 pm on Sep 9, 2008 (gmt 0)|
|Only because the webmaster tools give off a small warning when your description tags are duplicate? |
I think it's more than that. Pages without <TITLE> tags - or site wide identical titles - simply don't get listed as effectively as those with unique page titles. They fall into what used to be the 'supplementary' results, treated like duplicate content on the page.
Additionally, when the pages DO appear in serps, they are missing the opportunity that a good, unique title can bring.
Similar story with descriptions; they can damage indexing (I have first hand experience on that, as have many others around the threads here!), and also markedly increase the chance of an awful-looking serp that few would choose to visit.
Take webmaster tools' advice on this one ;)
[edited by: Quadrille at 1:40 pm (utc) on Sep. 9, 2008]
| This 41 message thread spans 2 pages: 41 (  2 ) > > |