For example, let's say you are a retailer who wishes to sell your products online. You only have one catalog but for ease of use you'd like to be able to present products in the localized currency and dialect of the customer. Obviously there are more similarities than differences in the English language whether you are from the US or Australia.

So how do you ensure that the search engines don't penalize you for duplicate content, yet still ensure that people outside of the US get to see the proper information? You use IP recognition.

There are 2 steps to this process. First is to ensure the search engine spiders only receive 1 version of the site - preferably the US English version.

You want to serve them this version because this is the version you want indexed by Google, Yahoo! and MSN. Granted this is also the version that will end up in the UK, Australian or Canadian engines because those versions of these sites share the same index. But that's ok because we'll deal with that in a minute.

What this does is forces the crawlers to see only one English version of the site despite the fact that you have an UK, Australian and Canadian version.

So by using IP recognition (and a list of spider IP addresses) you can ensure that all the main American spiders only see the American English site. And this is what you want.

The next step of this process involves ensuring that non-American visitors see the proper localized version of the site. This too involves IP recognition, but recognition of a different sort.

Using an IP-to-country database, your site can perform a lookup of an IP in a database and match it to a country. Then, using that information the site can serve the visitor the proper content, based on that match of IP to country.

And the best part of this is the IP to country database is free, and updated on a regular basis. If you go to this site you can get more information on implementing the IP-to-country database.

So let's take a look at the scenario introduced above - the vendor wanting to go global.

Now, rather than the appearance of duplicate content, there is only one version of the site which gets indexed - the version with US information and pricing.

A visitor uses a search engine, regardless of whether it is Google.com or Google.co.uk and performs a search. They come across the listing of the vendor and decide to check it out. The listing they see in the SERP is for the US version of the site; however when they click on the link the first thing that happens is that the visitor's IP is checked against the database and if they are from the US then they are served that page.

However, if they are from another country, such as Australia, they are automatically redirected to the appropriate corresponding Australian page without noticing that it has happened.

Now let's say that Australian visitor purchases the product and is so happy with it, they link to it on their site. But their link is to the Australian page because that's the page they were served. Then their site gets crawled by Googlebot, or one of the other crawlers and requests the Australian version of the site via the link on the customers site.

IP recognition is used and it determines that this is in fact a crawler, and it is then redirected to the corresponding US page.

Seems pretty simple huh?

Granted I've over simplified the scenario somewhat - there is some other work involved coding the pages to ensure that the visitor is sent to the proper region specific page, but in essence this is how the system should work.

The end result should be that the visitor's IP is recognized and they are sent to the proper geographic specific section of the site without any delay.

I can tell you that I've consulted on the set up of just such a system and it does work. I have a well known travel site as a client and this is the exact scenario presented to me. They have English regions other than the US which they service, such as the UK, Australia and Canada. And while the language is the same (except for minor differences) the travel pricing varies quite a bit.

With this system in place not only does the site visitor get served the proper content, but they are also served localized pricing which reflects the differences in cost to travel from one of these regions and not the US.

They don't have an issue with duplicate content, because the search engines only receive one copy of any given page - the US copy, yet when visitors are referred from Google, Yahoo! or MSN they always get the proper copy complete with proper pricing, regardless of where the search was performed.

So if you are considering going global with your product or service but are unsure how to move forward be sure to use IP recognition as part of your strategy. It can save you many headaches and much frustration with the search engines.


Rob Sullivan
Head Organic Search Strategist
Enquiro Full Service Search Engine Marketing


Copyright 2003 - 2005 - Searchengineposition Inc.