Search engines look for matches between keywords the user enters and words in the retrieved documents.

If you got any result for your query it means that the result is the right one and that there is such a match.

In everyday life when you communicate with human beings and ask a question - the moment you get the right answer you are satisfied and your quest is finished. With search machines getting the right answers might be only the beginning of the quest.

There are close to 135 million right matches for the words John Smith in Google. Putting these words in quotes reduces this number to 3,590,000. We assume that one or more of these are relevant but how should we find out?

People say that clustering helps in solving this problem. I typed the words "john smith" in Vivisimo and got the following list:

"John smith" (205)
Captain John Smith (50)
Pocahontas (23)
Colony (12)
New England (9)
Coach (7)
Scotland (7)
University (7)
Texts (7)
Collection (8)
Real estate (6)

I hope that this illustration helps you understand how complicated is this issue and what are we dealing with when we talk about relevance of search engines. All search engines are very good at finding unique words like "laurel leaves" and all search engines are helpless when they deal with problems like "John Smith".

Between these two extreme examples there is a gray area and that's where you can see which search engine is better than the others and in what feature. In this area QTSaver has two advantages that might help decide if you are on a wrong track:

a. The leads are a list of special words that give you an overview of the subject. If you are looking for a dentist and the leads are telling you about a flower shop it means that you need to refine your search.

b. The proximity of the paragraphs helps understand quickly where you are. In seconds you can read few paragraphs and see if they talk about your subject or not. In a regular search engine you need to read a few pages in order to arrive to the same conclusion.