Main menu • Search basics • ScreensKnow-howSubject Searches | Buy useSim

What is a keyword?

keyword: a significant word from a title or document used as an index to content.
The Webster's Collegiate Dictionary

The dictionary definition says it all: in order to be a keyword, the word must be significant, it must be in the text of a document, and it must point to the document. If you know how search engines index web pages, you'll understand each part of the statement. It is very briefly described in the next paragraph. Without getting too involved, it is enough to feel the connection between keywords and web pages they are in. Intellectually or viscerally, doesn't matter.

Search engines access websites, download web pages, and create databases consisting of all the words from downloaded texts. Each word in such a database is equipped with pointers to all the web pages where it was found, and with an indication of its position in each page relative to other words. It is the essence of indexing the Web. Any indexed word can serve as a keyword. When a user queries a database with a word, it returns the addresses of all the Web pages where the word can be found. The word becomes a key to the searched information. Or a keyword.

Any word can be used as a keyword, but it is not always practical. Some of the words in the text are very common. In fact, they can be found in any text, so they are of little use in finding texts by keywords. Therefore search engines ignore them. In the search engine jargon they are called "stop words". 

The list of stop words differs from one search engine to another, but the core of it is the same. It usually includes: a, an, and, any, as, but, by, did, do, from, he, her, how, if, in, is, it, its, no, of, what, when, you, and so forth. In addition to "classic" stop words many search engines ignore single letters, certain single digits and signs. The purpose of that is to speed up processing and save resources.

The uniqueness of a text is created by other, more meaningful words. It is such words that serve as keywords.

To illustrate it, we can s out from the text in the picture all the articles ("a", "an", "the"), as well as prepositions ("and", "or", "in", "on", etc.) and some other very common words.

Any of the remaining words in the text can become a keyword.

This is the raw material for the searcher to pick up keywords from.

Notes.
• Words in the graphic elements of a web page, such as "Stem Cell Information" in the picture, are not searchable.
• Sometimes you don't see your keyword in the found page. One reason may be that the search engine used it as
  a "cue word". For instance, if you search for "car dealer", the search engine may also return pages with "auto dealer".

In the web searching context. a keyword is any combination of letters, numbers, or other characters. Computer regards it just as a string of any characters between two space characters. When you send a keyword query to a search engine, the search engine computer finds a matching string in its database and returns the addresses (a list of links or URL's) of web pages having that string in their text. That is why any misspelled letter (character) kills the query. Just like ordinary words, keywords are often misspelled. According to some estimates, one in five queries is misspelled. It is one of the main reasons for unsuccessful searches. If you have no or too few documents in the results list, make sure your spelling is correct.

Another important aspect of keywords that can be used in web searching is their position in the web page. See "Web pages" in the left panel.