PDA

View Full Version : Chris: Articles post-Florida



pas
01-13-2004, 06:12 PM
Chris - In light of Google's latest changes (i.e. Florida), is there anything you'd change in your articles?

Chris
01-13-2004, 06:56 PM
No. Other than if I mention Google not stemming.

pas
01-13-2004, 07:02 PM
Have you figured out (or at least narrowed down) why Google drops sites for certain terms?

Chris
01-13-2004, 09:30 PM
I have theories but nothing I am going to put into an article yet.

IMO the most likely candidate is Google looking more at the subject of the sites that link to you (beyond just anchor text).

Mike
01-14-2004, 09:00 AM
What are your theories on anchor text Chris?

Chris
01-14-2004, 09:03 AM
I don't think anchor text usage has changed much at all.

Mike
01-14-2004, 09:14 AM
So by linking to your index page for example, from every page of your site using the same anchor text will be ok? I hope so...

chromate
01-14-2004, 11:51 AM
It hasn't done my carbohydrate counter site any harm. That's linked to with "carbohydrate counter" for every single link.

pas
01-14-2004, 12:28 PM
One thing I noticed when comparing my dropped pages to those which are still highly ranked as that the latter used "home", "index", etc. when linking back to their index (mine used keywords). Just a hypothesis at this point (anybody have URLs that prove it wrong?), but I changed one of my sites and will see if it makes any difference.

ozgression
01-14-2004, 04:37 PM
I think that the <title> tag and keyword richness of domain has been given more importance...

I think the importance of anchor text and PR has been downgraded due to the fact that they are very easy to manipulate.
________
Daytona (http://www.ferrari-wiki.com/wiki/Ferrari_Daytona)

DaveM
01-14-2004, 04:51 PM
^Exactly what I was thinking.

That's why from now on Im always making sure the domain I get has the keywords in it I plan to target....

Shawn
01-14-2004, 04:54 PM
I think PR isn't really a major deal at all anymore. At least it doesn't seem that way right now.

Dave, I PMed you.

Chris
01-14-2004, 05:17 PM
Originally posted by ozgression
I think that the <title> tag and keyword richness of domain has been given more importance...

I think the importance of anchor text and PR has been downgraded due to the fact that they are very easy to manipulate.

How easy is it to manipulate your title tag or domain?

Of all things, PR and anchor text are the hardest things to manipulate because they require either the help of another webmaster, or for you to have another site that was linked to by other webmasters.

Chris
01-14-2004, 05:55 PM
http://www.stuffyoumightlike.com/google-report-advisor.pdf

A good read, mostly same thoughts I have. Basically what you need is incoming links from sites dealing with your same topic.

Of course that hasn't been proven yet, it can't be since Google hasn't updated enough times to provide for adequate experimental data.

chromate
01-14-2004, 06:27 PM
Originally posted by pas
One thing I noticed when comparing my dropped pages to those which are still highly ranked as that the latter used "home", "index", etc. when linking back to their index (mine used keywords). Just a hypothesis at this point (anybody have URLs that prove it wrong?)

Again, my carbohydrate counter site uses "carbohydrate counter" as the link back to the home page from every page on the site.



Originally posted by Shawn Weeks
I think PR isn't really a major deal at all anymore. At least it doesn't seem that way right now.

I actually think PR is just as important as it always was, from what I've seen at least. Finding a site that has a high PR lower in the SERPs than we expect could merely mean that we don't understand how google are now interpreting the other aspects of SEO, that may actually be responsible for that page's lower ranking. I'm still seeing a definite correlation between PR and the sites returned in the SERPs. Got any examples? I only really monitor my keywords, so I'd be interested in taking a look.

ozgression
01-14-2004, 08:58 PM
Originally posted by Chris
How easy is it to manipulate your title tag or domain?

<Title> tag: Easy, but you have to choose ONE or TWO terms instead of many terms. That is, you cannot simply get all the SERPs you want for any keyword on your page because of your PR. I believe it is given more importance than PR and anchor text for this reason.

Domain: Well easy enough I guess. But i meant that a search for "cars" will bring up cars.com before myboringsiteforcars.com (this is taking PR and anchor text out of the mix in this example). Naturally, the better quality sites, usually have the more relevant domain names, thus making domain manipulation harder (although not impossible).


Originally posted by Chris
Of all things, PR and anchor text are the hardest things to manipulate because they require either the help of another webmaster, or for you to have another site that was linked to by other webmasters. [/B]

Wrong. You can buy a text link on a PR9 site for a few hundred dollars. I see threads/forums/sites everywhere trying to exploit this. Let alone sites that spam guestbooks/forums/blogs.

Also, instead of rewarding the most relevant sites, PR and anchor text rewards the most SEO savvy (ie. us :)). Therefore, I believe Google gave less importance to PR and anchor text. Not discounting them completly, but giving them far less importance than before.

Oddly enough, simply keeping the Google PageRank from displaying on the toolbar would get rid of the problem of PR manipulation. Displaying PR has no use at all to web surfers.

Anyway, my opinion is just that, my opinion. I am not passing what I say off as facts...

cheers
________
Honda Atlas Cars Pakistan specifications (http://www.honda-wiki.org/wiki/Honda_Atlas_Cars_Pakistan)

Chris
01-14-2004, 09:34 PM
Wrong. You can buy a text link on a PR9 site for a few hundred dollars. I see threads/forums/sites everywhere trying to exploit this. Let alone sites that spam guestbooks/forums/blogs.


How is this different from:



Domain: Well easy enough I guess. But i meant that a search for "cars" will bring up cars.com before myboringsiteforcars.com (this is taking PR and anchor text out of the mix in this example). Naturally, the better quality sites, usually have the more relevant domain names, thus making domain manipulation harder (although not impossible).


People who can afford the higher quality domains tend to have higher quality sites. People who can afford to pay for text links are the same people who can afford to pay for nicer domains.

You can simply divide it into those who have money and those who do not. Those with money can afford to pay better designers, they can afford to pay employees, they can afford to pay for content, they can afford to pay for domains, and they can afford to pay for links.

If it is your opinion that people with money tend to have better sites then buying text link advertising isn't so bad.

PR can definitely be manipulated, but it is harder to manipulate that just about anything else. It doesn't cost me a few hundred dollars to manipulate my title tag.

Also title tags are not given more important than PR/Anchor Text and the "one or two words" thing is irrelevant. First of all you have more than enough room for a good 4 or 5 words if you want them, the keyword being if. Most people focus mainly on a single word or phrase for each page. There will be secondary keywords to be sure but only one main one.

Even considering that I think if you checked most page titles are longer than the incoming links that point to those pages, as such anchor text provides less keyword room than titles.

chromate
01-15-2004, 05:23 AM
There's something wrong with saying "anchor text is more important than PR." I don't think it's correct to even compare them in the first place. This is because there's no constant that's common between them. It's like saying "this is a long path". Well, compared to what? PR is less important... Compared to what? Anchor text? It just doesn't work because they're hardly related.

For example. A PR10 site is useless until you lend context to the PR through anchor text and the like. It's hard to explain, but I always think of PR as "empowering" the content you do have.

Best.Flash
01-15-2004, 01:56 PM
Originally posted by Chris Other than if I mention Google not stemming.

It looks like your saying here Google have stopped stemming post Florida?

flyingpylon
01-15-2004, 02:44 PM
I don't mean to speak for Chris, but I took that to mean that if he mentioned in the article that Google does not do stemming, that he would need to update the article to say that Google does do stemming now. (sorry if I'm wrong about that Chris)

Chris
01-15-2004, 03:34 PM
You're correct.

snoopers
01-18-2004, 10:23 AM
Chris, thanks for the link.

How could the theory apply to http://www.tropicalresorts.org/?

All of its backlinks come from a physics site, yet it is ranked 2nd for the search term "tropical resorts" (which got 73k searches according to Overture)!

Shawn
01-18-2004, 10:29 AM
Tropical Resorts isn't a popular search term. It receives less than 50 searches a day.

snoopers
01-18-2004, 10:37 AM
Which tool are you using?

Try this

http://inventory.overture.com/d/searchinventory/suggestion/

chromate
01-18-2004, 10:40 AM
try wordtracker.com ... it tends to be more accurate.

snoopers
01-18-2004, 05:59 PM
Wordtracker only uses results from Altavista.

Overture is very accurate, because its their direct logs.

ozgression
01-18-2004, 06:54 PM
Try this tool... http://www.digitalpoint.com/tools/suggestion/

Search both :)


From their site
Why do they differ so greatly? Overture's data is based on more searches, so the values are going to be higher than Wordtracker's in most cases. This does not mean Wordtracker's data is less valuable (keep in mind that they are relative to each other). Also, Wordtracker differentiates between plural and singular forms (which Overture does not).
________
Suzuki RGV500 (http://www.suzuki-tech.com/wiki/Suzuki_RGV500)

chromate
01-19-2004, 03:55 AM
Originally posted by snoopers
Wordtracker only uses results from Altavista.

Overture is very accurate, because its their direct logs.

That's precisely why it's not very accurate :) Wordtracker only uses altavista in the trial version. In the paid version you get a nice selection of search engines to compare stats for. It's easy to predict what the google result would be based on the Altavista result anyway.

flyingpylon
01-19-2004, 06:49 AM
Originally posted by chromate
It's easy to predict what the google result would be based on the Altavista result anyway.

Is there a specific multiplier that you use when comparing results between the free version of Wordtracker or Overture and Google?

chromate
01-19-2004, 07:50 AM
Not an exact multiplier. But as a guide 50 searches in 24 hours on altavista converts to about 1000 on google.