An apparatus, program product and method utilize automated analysis techniques to assist in the determination of acceptable usages of linguistic terms (e.g., words, phrases, acronyms, etc.). In particular, an acceptable usage of a linguistic term may be determined by scanning a plurality of documents for variants (e.g., based on differing spellings, punctuation, capitalization, meaning or definition, etc.) of the term, and tracking relative occurrences of a plurality of such variants found in the plurality of documents during scanning. By tracking occurrences of linguistic term variants, users may be able to use such statistical information to select which of the available variants represents an acceptable usage, or even a most acceptable usage, of a term. Scanned documents may be retrieved from the Internet, and scanning may occur while a user is browsing the Internet.