Jump to content

  • Log in with Facebook Log in with Twitter Log In with Google      Sign In   
  • Create Account

Subscribe to HRA Now!

 



Are you a Google Analytics enthusiast?

Share and download Custom Google Analytics Reports, dashboards and advanced segments--for FREE! 

 



 

 www.CustomReportSharing.com 

From the folks who brought you High Rankings!



Photo
- - - - -

Keyword Phrases


  • Please log in to reply
48 replies to this topic

#31 Randy

Randy

    Convert Me!

  • Moderator
  • 17,540 posts

Posted 02 June 2009 - 02:41 PM

I'm not flaming you adibranch. Trust me, you'd know it if I were. giggle.gif

Seriously, I'm just challenging your opinion with actual test data so that others reading along --who seem to too often read something on the 'Net and trust it blindly-- have both sides of the argument. No hard feelings at all on my part, of which you can be 100% sure. Fact is I like people who push against the status quo. That's how we learn more. wink1.gif

And for the record if you think the client site example is a test, it's not. (And yes I know you didn't say it was your test, but you're offering it as evidence of your position.) What you describe could have been a test, but it's not as you described it. It's simply empirical evidence of a single case that might or might not mean much of anything.

If you had wanted to really test it you could have changed it from all uppercase to lowercase or mixed case, saw movement, then changed it back to see if the results moved again, then changed it again to see if the results moved again... Then do the same thing with 20 or 30 other examples and different phrases. With enough matching data you might have been able to say with a reasonable degree of certainty that you were onto something.

That's the thing with creating valid tests. They have to be repeatable. Repeatable by you and by others. Doing something once doesn't make the outcome, no matter what the outcome is, very reliable. It just gives one a starting point that is based entirely upon single source empirical evidence. And this how way too many people end up mixing up cause and effect.

I also understand that one usually doesn't want to do this sort of thing with the sites of paying clients. That's why for years I have maintained a healthy group of test bed domains. Though I do also test a lot of stuff on my real sites, that's a different story than most. They're my sites, so it's totally up to me whether the potential value of gaining the information is worth the potential short term losses of testing with an income producing site.

#32 adibranch

adibranch

    HR 5

  • Active Members
  • PipPipPipPipPip
  • 332 posts

Posted 03 June 2009 - 01:16 PM

okay fair enough.. biggrin.gif we'll draw a line under this one and leave it at that.

Although, i did come across some blog stuff recently stating that case is being used by google for different things.. ill dig them out and post a link.

#33 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 04 June 2009 - 04:23 PM

Since this case-sensitive discussion began I decided to see where the thought originated. I've read several UK SEO bloggers (who I won't mention) writing about it recently. Seems case-sensitivity has been more noticed in the UK than in the US. Using my kwd lists I can't get the SERP listing to change. However, using these phrases [fossil watches, Fossil Watches, Chicago Hotels, chicago hotels] the SERP does change slightly. The funny thing is each SERP listing has the same case. So why the movement in rankings? I also don't see how one could assume with such a very small change in the SERP listing that it means G is doing case-sensitive weighting. They still say case doesn't matter on their site: G Case Statement. It's more likely an algo issue rather than a plan to start using case. In addition, their auto-complete feature (which they keep enhancing) uses only lower-case. I could be wrong, but either way I now have something else to put on my plate to observe over the course of the year. I'm not changing anything on my sites and it'll take a lot more than one position move in the SERP to convince me that it's something worth taking seriously.

#34 adibranch

adibranch

    HR 5

  • Active Members
  • PipPipPipPipPip
  • 332 posts

Posted 05 June 2009 - 04:28 AM

my original point was that capitalisation of first letters of keywords makes a possible difference due to its natural use in grammar.. it would seem that your test may back that up to a very small degree. And recently, since i've been looking into it, i've found reports that it is taken into account in various google algo, but mainly relating to adwords.

#35 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 05 June 2009 - 08:34 AM

I haven't seen any reports about weighting case relating to Adwords.

Again, I haven't seen a US SEO blog indicate what you Brits are saying. BTW, each of the three bloggers I did read used the [fossil watches] as a test (it's not my test). The third blogger found [chicago hotels]. Strange they couldn't find more examples to make their case [pun intended]. Again, out of several hundred kwd phrases in my industry the SERPs do NOT change. In addition, the movement of 1 position in the middle of the top 10 SERP is not something that should give you or anyone else confidence that G is weighting case. Especially, since each site that was swapped used the same case in the title of their pages. Lastly, it should give anyone cause for pause that G uses lower-case for their auto-complete stuff - which builds a much stronger position that lower-case is preferred to proper-case. Also, I'm not leaving out the possibility G could be messing with the UK by testing a case-sensitive algo there. That's feasible, but to make broad stroke statements that case weighting is going on, because it's a logical thing to do, is an extremely weak position with all things considered.

adibranch I'd still be interested in trying some of the tests you used. I'm a test junky.

Edited by BBCoach, 05 June 2009 - 08:43 AM.


#36 adibranch

adibranch

    HR 5

  • Active Members
  • PipPipPipPipPip
  • 332 posts

Posted 05 June 2009 - 01:23 PM

again, it is an opinion, but it makes sense. It states in the adsense guidelines not to use excessive capitlisation to try and gain attention to your ads, or the ad will be removed. I'm just translating that into organic listings too... or, we'd all be doing it wouldnt we with no repurcussions whatsoever. But, i have no evidence that this is the case.... yet.

The other issue of capitalisation of first letters is a seperate one. I can only suggest that capitlisation occurs naturally in copy for word reinforcement, names, products, brands etc, and may be therefore an easy way for google to pick out keyterms from text. Either way, it occurs in most of my copy as a matter of course, for both grammatical reasons and for possible SE reasons .

But, like i say , this is opinion, its open for discussion as far as i'm concerned.

#37 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 05 June 2009 - 02:00 PM

QUOTE
adibranch - I have tested, in the same way you have.. so we got different results, but that just means there are too many variables involved. On here, I offer opinions based on findings, and i state where possible that thats is exactly what they are.
Ok. So it's open for discussion, but do you have any tests that I could play with to further test this?

#38 Randy

Randy

    Convert Me!

  • Moderator
  • 17,540 posts

Posted 05 June 2009 - 03:10 PM

It's also important to remember that Adsense/Adwords and Organic are completely separate entities and separate business units that don't even talk much to each other within Google. You really can't take even confirmed evidence from one side of this Paid or Organic equation and apply it to the other. They're different animals and in many, many ways are treated as totally different animals. Beginning of course with how one gets a PPC ad to be placed highly as compared to how to get a good organic ranking.

Just a reminder that it's dangerous to make any such assumptions.

#39 adibranch

adibranch

    HR 5

  • Active Members
  • PipPipPipPipPip
  • 332 posts

Posted 06 June 2009 - 05:10 AM

i think 'dangerous' is a bit dramatic..
sod it, i'm going to risk it.. i'm going in.... wish me luck.

#40 Randy

Randy

    Convert Me!

  • Moderator
  • 17,540 posts

Posted 06 June 2009 - 08:13 AM

I think you may have misconstrued what I meant adibranch. I don't mean it's dangerous in the sense that the site will be penalized or anything long those lines. I just don't see that happening.

What could be dangerous is to take something you see maybe, possibly making a difference in PPC and trying to extrapolate that it might also make a difference in organic. The two systems are just too different to make such leaps of faith.

#41 adibranch

adibranch

    HR 5

  • Active Members
  • PipPipPipPipPip
  • 332 posts

Posted 07 June 2009 - 06:07 AM

the PPC capitlisation issue i referred to is in googles adwords guidelines mate, its not a maybe possibly. Granted that it has no relevance to organic, i was just stating an example of automated case checking by google. I thought i was obvious about that.

With regards organic, as i've expained many times in this thread... capaitlisation of first letters occurs naturally in copy for keyterms.. theres nothing to lose.
If it makes a difference in algo's, i'm laughing.
If it makes it easier to the visitor when scanning copy, i'm laughing.
If it does neither, slap me for being grammatically correct and making the site look like its had a modicum of thought put into it.

I cant see what the problem is. I did, after all, only suggest it as good practice, nothing else.

Edited by adibranch, 07 June 2009 - 06:13 AM.


#42 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 07 June 2009 - 08:31 AM

One question, adibranch...

How does capitalizing a keyword phrase make it grammatically correct? Unless you're talking about a proper noun, most keyword phrases should not have initial caps. A proofreader or editor would fix it to all lower case.

#43 adibranch

adibranch

    HR 5

  • Active Members
  • PipPipPipPipPip
  • 332 posts

Posted 07 June 2009 - 09:36 AM

now you're just being pedantic biggrin.gif and were getting away from the issue which was being discussed.
okay.. branding, brandnames, product names, nouns relating to the subject matter or their semantics. Fair enough? I'm not a proofreader...

#44 seoseattle

seoseattle

    HR 1

  • Members
  • Pip
  • 9 posts

Posted 07 June 2009 - 11:43 AM

GREAT THREAD! eek.gif

My experience with keyword phrases is that I try and keep the keyword limit to three with the exception of adding a GEO. For example - Janitor Cleaning Service Seattle -. The only important universal 100% valuable consideration that should be made in each and every selection of a keyword phrase is to remember that Google puts more weight on the first word in a keyword string both on Query and Search Return. I would put the word with the highest search volume first. EXAMPLE: - Birthday Gifts Personalized - and not - Personalized Birthday Gifts -

#45 mal4mac

mal4mac

    HR 6

  • Active Members
  • PipPipPipPipPipPip
  • 610 posts

Posted 07 June 2009 - 12:03 PM

QUOTE(seoseattle @ Jun 7 2009, 12:43 PM) View Post
Google puts more weight on the first word in a keyword string both on Query and Search Return. I would put the word with the highest search volume first. EXAMPLE: - Birthday Gifts Personalized - and not - Personalized Birthday Gifts -


Do a search on your example. No proof of your "word with the highest search volume first" thesis there! Quite the opposite...

If "Personalized Birthday Gifts" sounded better I would use that initially and see if it scored well for "Birthday Gifts Personalized"

If it did not achive the SERP I required I then might experiment with changing it to ""Birthday Gifts Personalized". But I'd do other tweaks first!

Also even if your trick worked what if "Personalized Birthday Gifts" had only a slightly lower search volume but much less competition?





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

We are now a read-only forum.
 
No new posts or registrations allowed.