Handling the problem of finding all of the occurrences of a string in a given text is solved through using string matching techniques where string matching represents one of the major problems that should be solved by computer technologies. With regards to information retrieval, pattern matching was widely implemented as well as Web search engine. There are major and important tasks executed by string matching among it the following jobs: signal processing, speech and pattern recognition, library systems and web search engines as well as information retrieval. This paper proposes a new implementation in string matching to improve the search technique which takes advantages from building a table consists of two columns, the first one is the words lengths exists in the text and the second one is the start positions of each word classified by the same length. The algorithm performs preprocessing for the pattern and for the text before beginning to search.