Forum Moderators: open
[chat.abondance.com...]
Jean-Delatour : Is it still necessary to fill up the meta-keywords ?
I think that we're open to either kind of deal. Let's see. Google uses meta-keywords, but not as much as most other search engines. I would still include them, but don't worry about putting a ton of effort into it.
Calum has said that he thinks that they only use them if there isn't anything else to go on on the page (I think that is probably correct).
I don't usually use them, and when I was doing a lot of research - I did not see that they had any real effect. I don't think the extra 60 seconds is usually worth it to put them in.
I think that Calum is probably right - or it adds something like 1% help.
If that is the case and you want to use them - I think google could care less if you use commas or not - I can't imagine that would make a difference in more than 1 out 100000 cases. I would use the commas.
As a standard, I always put meta tags. Regardless of the big SE's, many a smaller site may have their own free search software that picks up your description and keywords. If you dont have them in...well....you might not rank on 101 like minded sites :)
Only negative about having them is bandwidth and code/text ratio....but I always put in 5/15 keywords anyways :)
I like my competitors to use META tags, though. When I start to think about a new site, the existing competition's METAs are some of the first places I look when I start to build a 'seed' list for keyword research.
I think the "no comma" myth started when people were keyword stuffing the metas and didn't want to waste valuable space on a lowly comma. Now that lean and mean keyword metas are in vogue, I've seen more SEOs using commas.
The last few sites I've optimized, I kept the keywords grouped in natural phrases, used a comma to separate them and only optimized for two phrases per page. Inktomi engines seem to like that title, meta, header, anchor, body phrase if it finds the same phrase in all 5 spots, much the same way Google does. If I'm cloaking the site, I do create different titles for Inktomi engines but keep the meta keywords separated by commas. I simply don't use any keywords that aren't in the title of the page. Keeps the page focused, and makes it much easier to determine the anchor text for inbounds and internal links.
Stripping the commas out of the metas to see how Inktomi reacted was inconclusive, sites rose and fell in the serps regardless of whether commas were used or not.
As for experienced SEOs NOT using commas, well, experience comes from testing. Unless you are certain that the person promoting one technique over another has tested that technique in a multitude of ways, weighed the variables and then reached a conclusion based on sound analysis, discounting or accepting that view is inherently risky. Not so much in this case as it is meta tags we're talking about.
You can isolate all the factors used in your optimization, but you can't isolate what your competitors have done or changes in the engine's algo so even careful onsite optimization analysis needs to be weighed before you can say with certainty that a particular technique works.
I've seen sites doing quite well with, "put keywords here" in their meta tags...Maybe it's the lack of commas in that phrase that gives them their boost. ;)
DG
Without the comas the engine is free to form any combination of your keywords into a keyword phrase
That was one of the theories. I use focused keyword phrases and don't need any help from the engines determining what is relevant. If there are too many phrase options, another page is needed. If the phrase is focused, rearranging it is senseless. :)
Edited to fix the quote tag
DG
If one page has a keyword phrase "bible questions and answers," is it not optimized for that specific word order? But if instead the meta tag listed "bible questions answers" (no coma) would it not be equally be optimized for search strings of "bible questions answers" or "bible answers questions" or "questions bible answers"? Or is this just like trying to count how many angels could fit on the head of a pin?
When I was an About.com guide, About.com's SEO guru advised us *not* to use commas, and I've also heard that advice from others who have been in the search business for quite a while. The theory seems to be that commas prevent the search engine from recognizing word combinations. In other words, if your keywords read "London, Chelsea, hotels" and somebody searches on "London hotels Chelsea," the commas will keep the search engine from assembling that phrase from your keywords.
In practice, meta keywords aren't nearly as important as they once were, so the comma-vs-no-comma question may be academic.
My keyword tags are very short, and though I don't generally repeat keywords in the tag to get different phrase matches, I do try to get my most important phrases together early on in the tag. In the larger scheme of things, I don't think the keyword tag is very important, so I don't depend on it, but I try to be careful just in case it is. You might say I'm a keyword tag agnostic.
This test isn't very scientific, as search box syntax would also be involved, but, trying some very quick searches on Google, AV, Fast, and Ink for <keyword1,keyword2> and <keyword1 keyword2>...
- on Google and Ink, the comma doesn't seem to make any difference.
- on Fast and AV it does.
Because of possible search box syntax, this may not tell me much about how Fast and AV view commas in the keywords tag... and Google doesn't use the keywords tag... but my sense of consistency does suggest that if Ink ignores commas in the search box, it would ignore them in the keyword tag too.
I seem to remember that Brett suggested an approach to the keyword tag that's stuck in my mind ever since. I believe that what he suggested was to choose a handful of keywords that define the primary themes of the site or of the site section, and use those on every page. The objective would be to help themes-based engines establish the overal themes for the site.
Then -- he suggested that you add another handful of keywords that are specific to the page. The end result is that each page has a keyword tag that combines overall site themes with page-specific phrases.
The only thing that I've always wondered about with this approach is the fact that by using the site themeing keywords I'll frequently run into situations where they do not match specific page content, and for one reason or another I cannot alter the page text. If I'm to believe some of the other "conventional wisdom" (which is further perpetuated by automated optimization tools), it's a "bad" thing to have keywords in the meta tag that are not conatined in the text on that page.
There's a part of me that prefers the logic that I've attributed to Brett. On the other hand I'd like to avoid being penalized for keywords that do not appear in the page text.
Anyone have any thoughts, or is this not even worth obsessing about?
Yet, when doing a paid Ink submission at one of the partner sites it says to use the commas, and suggests a limit of 255 characters. There's also a suggestion not to use any keywords that aren't on the page.
Google doesn't use what's in the description meta tag for their description, which is better, imho, at least it's an attempt to determine context. But I don't know if it's possible to figure whether either of the meta tags could count toward the keyword density. I'm checking density with and without just to be sure, but keep metas very short anyway - when they're used.
>can't see how it makes any difference whether the delimiter is a comma or a space.
That makes sense. In the days when people stuffed the keywords tag (some still do) the commas could have left less room for a long laundry list of keywords. If there's a maximum number of characters allowed, that's also a consideration.
Personally, I don't pay that much attention to the keywords tag. Trying to keep 3 words between repetition of any word with a maximum of using any word 3 times, using only a few variations of phrases, to me the commas don't make much difference. I use so few words, to avoid dilution, it's no difference. I'm of the no comma school, but I've started to use them some for some Ink submissions.
Even though I've paraphrased Brett on a possible approach to the keyword tag, he's also one of the first to tell you that meta keywords are probably not worth the extra bandwidth!
I guess one of the reasons I still feel compelled to use them is that it's hard to get past client perceptions that they're important and, as someone suggested in another thread, you can always point to them later if the client suggests that they might not be ranking well because you didn't use them (even if it has nothing to with their ranking).
(edited by: visibot at 1:08 pm (utc) on April 18, 2002)
If they still don't get the point, see about getting out of the arrangement. Clients like that (a little knowledge...) are often more trouble than they could ever be worth
It's worth doing, as TallTroll suggested, using that approach, and billing for the time. There are still a lot of people out there who still think that's what gets rankings, and also a lot who buy into getting listed through "optimizing submissions" and submitting monthly. They pay for that, sometimes a lot. And when we say otherwise, some have heard it so much that we risk losing credibility.
>>hard to get past client perceptions that they're important and, as someone suggested in another thread, you can always point to them later if the client suggests that they might not be ranking well because you didn't use them (even if it has nothing to with their ranking).<<
Exactly. So do we tell them the way it is, or what they've come to believe and want to hear? If they also believe that resubmitting monthly is what helps and we know we don't need to, do we say that, or offer monthly submissions as a service? Will people think we're doing less than the meta-tag/submissions crowd?
(edited by: Marcia at 12:58 pm (utc) on April 18, 2002)
I try to use the process of coming up with a focused list of search targets as part of the early communication process with the client... about search priorities, what can be attained and what can't be, and site focus. This is helpful throughout the process of optimizing the site.
So I'm really creating the keyword tag anyway... and a short one at that... as I go along. Might as well include it.
For a thread about meta keywords and page copy, which is where some of these posts are drifting, also see:
[webmasterworld.com...]