Search engines prominently use inverted indexing technique to locate the Web pages having the keyword contained in the users query. The performance of inverted index, fundamentally, depends upon the searching of keyword in the list maintained by search engines. This paper presents a new technique for keyword searching. It uses a trie data structure to index the keyword up to a certain optimum level. While searching a keyword, this index is used to get two offset values, in constant amount of time for every keyword, within which the keyword might lie. Using the two offsets, a binary search is initiated to locate the keyword in the list, and hence the Web pages containing the keyword. Research shows that subsequently increasing the levels of trie will increase the performance of retrieval but also increase the required memory. It also shows that on an average with indexing up to level 2 requires 56% less number of comparisons, as required by binary search, to search a keyword in the list.