thinks he'll have to go back and differentiate these methods again in the future, changing the calls to the method is probably not much more difficult than a search and replace in the calling code blocks. If he has already bounds-tested and otherwise proofed the data elements, a more generic insert/update method is called for-- and it will be faster to change the calling code than to refactor a long list of unique methods to call the generic method. If he has *not* proofed his values, though, he might want to leave these routines with unique names so that he can validate at this point-- which is probably too late {grin}.