Creating a robots.txt file
By Sumantra Roy

Some people believe that they should create different pages for different search engines, each page optimized for one keyword and for one search engine. Now, while I don't recommend that people create different pages for different search engines, if you do decide to create such pages, there is one issue that you need to be aware of.

These pages, although optimized for different search engines, often turn out to be pretty similar to each other. The search engines now have the ability to detect when a site has created such similar looking pages and are penalizing or even banning such sites. In order to prevent your site from being penalized for spamming, you need to prevent the search engine spiders from indexing pages which are not meant for it, i.e. you need to prevent AltaVista from indexing pages meant for Google and vice-versa. The best way to do that is to use a robots.txt file.

You should create a robots.txt file using a text editor like Windows Notepad. Don't use your word processor to create such a file.

Here is the basic syntax of the robots.txt file:

User-Agent: [Spider Name]
Disallow: [File Name]

For instance, to tell AltaVista's spider, Scooter, not to spider the file named myfile1.html residing in the root directory of the server, you would write

User-Agent: Scooter
Disallow: /myfile1.html

To tell Google's spider, called Googlebot, not to spider the files myfile2.html and myfile3.html, you would write

User-Agent: Googlebot
Disallow: /myfile2.html
Disallow: /myfile3.html

You can, of course, put multiple User-Agent statements in the same robots.txt file. Hence, to tell AltaVista not to spider the file named myfile1.html, and to tell Google not to spider the files myfile2.html and myfile3.html, you would write

User-Agent: Scooter
Disallow: /myfile1.html

User-Agent: Googlebot
Disallow: /myfile2.html
Disallow: /myfile3.html

If you want to prevent all robots from spidering the file named myfile4.html, you can use the * wildcard character in the User-Agent line, i.e. you would write

User-Agent: *
Disallow: /myfile4.html

However, you cannot use the wildcard character in the Disallow line.

Once you have created the robots.txt file, you should upload it to the root directory of your domain. Uploading it to any sub-directory won't work - the robots.txt file needs to be in the root directory.

I won't discuss the syntax and structure of the robots.txt file any further - you can get the complete specifications from http://www.robotstxt.org/wc/norobots.html

Now we come to how the robots.txt file can be used to prevent your site from being penalized for spamming in case you are creating different pages for different search engines. What you need to do is to prevent each search engine from spidering pages which are not meant for it.

For simplicity, let's assume that you are targeting only two keywords: "tourism in Australia" and "travel to Australia". Also, let's assume that you are targeting only three of the major search engines: AltaVista, HotBot and Google.

Now, suppose you have followed the following convention for naming the files: Each page is named by separating the individual words of the keyword for which the page is being optimized by hyphens. To this is added the first two letters of the name of the search engine for which the page is being optimized.

Hence, the files for AltaVista are

tourism-in-australia-al.html
travel-to-australia-al.html

The files for HotBot are

tourism-in-australia-ho.html
travel-to-australia-ho.html

The files for Google are

tourism-in-australia-go.html
travel-to-australia-go.html

As I noted earlier, AltaVista's spider is called Scooter and Google's spider is called Googlebot.

A list of spiders for the major search engines can be found at http://www.jafsoft.com/searchengines/webbots.html

Now, we know that HotBot uses Inktomi and from this list, we find that Inktomi's spider is called Slurp. Using this knowledge, here's what the robots.txt file should contain:

User-Agent: Scooter
Disallow: /tourism-in-australia-ho.html
Disallow: /travel-to-australia-ho.html
Disallow: /tourism-in-australia-go.html
Disallow: /travel-to-australia-go.html

User-Agent: Slurp
Disallow: /tourism-in-australia-al.html
Disallow: /travel-to-australia-al.html
Disallow: /tourism-in-australia-go.html
Disallow: /travel-to-australia-go.html

User-Agent: Googlebot
Disallow: /tourism-in-australia-al.html
Disallow: /travel-to-australia-al.html
Disallow: /tourism-in-australia-ho.html
Disallow: /travel-to-australia-ho.html

When you put the above lines in the robots.txt file, you instruct each search engine not to spider the files meant for the other search engines.

When you have finished creating the robots.txt file, double-check to ensure that you have not made any errors anywhere in it. A small error can have disastrous consequences - a search engine may spider files which are not meant for it, in which case it can penalize your site for spamming, or, it may not spider any files at all, in which case you won't get top rankings in that search engine.

An useful tool to check the syntax of your robots.txt file can be found at http://www.tardis.ed.ac.uk/~sxw/robots/check/. While it will help you correct syntactical errors in the robots.txt file, it won't help you correct any logical errors, for which you will still need to go through the robots.txt thoroughly, as mentioned above.

Article by Sumantra Roy. Sumantra is one of the most respected search engine positioning specialists on the Internet.

For more advice on how you can take your web site to the top of the search engines, subscribe to Sumantra Roy's FREE newsletter