SEO (Search Engine Optimisation)

Optimus01

Well-Known Member
Joined
Nov 16, 2009
Messages
127
Dude, you're going to wear out your thumbs.
Koedoes to Optimus01 for jumping in, head first.

I have 2 things I am going to ask:

I have heard, about SEO, that:
Your content must be mono-topical.
i.e.:
If your website covers too much of a range of stuff, it gets treated like a news site, and only the newest stuff is kept on top of the results. But, if there is very specific focus, you're already 70% of the way to being on top and staying on top.

Am I hearing wrong?

and

Google is localising results more and more.
i.e.:
Search results from a South African user tend to show results from local (South African) websites (locally managed, not locally hosted) with higher priority.

Care to comment on the viability of my observation?

From what I have seen, you are correct on both accounts.
 

convalescent

New Member
Joined
Dec 1, 2009
Messages
8
Wow, you do SEO and Search Engine Optimization?! That's... just... incredible! :eek:

whoa guys can't believe you flame a noob before he has even had a chance to explain himself. nice way to welcome him into the community.

Caffeine is something that I think every SEO has been watching carefully.

I don’t know exactly what the actual changes will be, as Google is holding this one close to its chest, however the talk is:

That is will be faster (when we talking milliseconds not a big difference) but still this is good.
More relevant (not too sure how as I find it pretty good at this now).

I have also read on a few sites that it will place more emphasis on real-time search so things like Twitter, Digg, Delicious and Facebook and the content therein will start affecting search.

I would suggest keeping your onsite stuff the same but maybe increasing your tweets

I would be interested to see if anyone else knows anything else???

yes and no.

no Google isn't really holding it THAT closely to their chest. Cutts seems to have been pretty open about what developments have been going on.

its not going to be a change in algo as it's more an infrastructure upgrade. shaving milliseconds off is damn impressive (btw if anyone is wanting to see the first roll out to a data centre you can find it here: 209.85.225.103) i'm seeing consistently about .3 to .2 milliseconds being dropped the information retrieval times

as far a relevancy algos go. Google don't do big announcements when it comes to algo tweaking. think Florida and more recently Vince (aka the brand update) where without warning the algo changed. infact the later was only picked up a month after it's roll out in feb of this year.

and the reason i'm saying this looking within my vertical on some core keywords i'm not seeing a change in SERP positioning across the pre-rollout data centre and the .com version (yes personalization is turned off)

real time search will become a more prevalent, but i don't see it being a direct influence on caffeine. also bing and google have only recently signed a deal with twitter to start incorporating twitters results in the SERPs

Why does Google index my affliate sites direct links?

For example, I'm affiliated with a gaming company that sells CD Key's online.

My sites structure links back to my affiliate site ID, for example: mygamingsite.com/instant-world-of-warcraft-cd-keys/

So someone will hit my site. Click on that link, and within that page there would be a "Buy now" link that looks like this:

mygamingsite.com/buy-now/instant-world-of-warcraft-cd-keys

This would then redirect to the affiliate gaming site with my affiliate Id

However, I found several times in Google that when someone searches for cd keys, and they find my site (name and everything) the link links directly to the affiliate site.

I'm not complaining, it has my affiliate ID in there and everything so it doesn't even hit my website (which is a shame because I'd like to assess traffic trends etc), but it's either indexed their site with my affiliate ID, or indexed my website content but linked directly to their site with my affiliate ID.

Any ideas? Good thing? Bad thing?

I forgot to mention that all my links for buy now does have nofollow on them... hence my question and why it was weird seeing Google index it...

Seriously, have you tried adding rel="nofollow" to your affiliate links? It prevents Google from indexing the link...

the more efficient methodology especially if you have those links in centralized "go" folder would be to use the robots exclusion protocol.

Code:
user agent: *
disallow /buy-now/

yes you can use the noindex meta tags but you're using the wrong method (well IMO).

because:
  • it's can be time consuming tagging those pages with the right nofollow / noindex directives.
  • you're leaking your own sites link equity.

(which is a shame because I'd like to assess traffic trends etc)

of course you can check the click through rates and traffic and maximize your ROI.

if using Google analytics add the following to the link:

Code:
onclick="pageTracker._trackPageview('/buy-now/wow-1');"

unfortunately this can also be a time consuming exercise and is kinda counter productive to the tip i gave above WRT to the nofollow.

my heavy duty / redundant solution is this:
  • i link to my "go" affiliates folder using a custom url shortener (which counts click thorughs) sitting on a domain i use exclusively for tracking and serving static data out to my sites.
  • the real URL (whiach has been shortened) has google analytics campaign tracking data attached to it
  • the redirects on that domain have a java scripted 302 delay on it to give analytics time to register the hit and the campaign data.

that whole (sub) domain does not get indexed which conserves link equity and gives me a pretty robust redundant system for gathering information on your demographics etc.

Seriously, if you know how to use standards to your benefit, you can change a website's coding setup to be more spider friendly.

indexability is only one part of the equation. yes good informational architecture plays a big role in how well you perform on the on page side of things. but stating that being standards compliant is going to give you a boost in teh SERPs is bogus information.

all things being equal it's your domains authority that will decide if your 1st or 2nd not whether or not you decided to make your site standards compliant. and i'm only saying this because myself {and other seos} have personally experimented with this supposed part of the algo and as far as i can has been debunked. as long as googlebot can access the information you're fine

That's how good companies make their money.

good search companies don't "make their money" by focusing on onpage factors alone. i've crushed sites in the serps who haven't focused on the holistic side of things and who only look at on page

Your content must be mono-topical.
i.e.:
If your website covers too much of a range of stuff, it gets treated like a news site, and only the newest stuff is kept on top of the results. But, if there is very specific focus, you're already 70% of the way to being on top and staying on top.

if that were true... why does Wikipedia rank so well?

google ranks pages not sites.

Search results from a South African user tend to show results from local (South African) websites (locally managed, not locally hosted) with higher priority.

yes google does give bias to ZA websites if you are using google.co.za ccTLD do give you bias but i'm seeing more and more bias being given to ccTLD and hosting in that country. the ccTLD isn't enough by itself. i'm so confused when people are trying to rank in the .co.za when there server is sitting in chicago?!?

practical example. one of our sites was developed for the Australian market. as a control the site was sitting in Canada sitting on a .com for 3 months, as said it was a control to measure regional bias.

last month i 301d it from the .com to a com.au sitting on an australian IP 1 month into the comparative test i've seeing a jump of close to 40% on organic search referrals.
 
Last edited:
Top