In general, formatted Excel spreadsheets have blanks somewhere in their structures that won't be blank in "true" data rows. The way I usually do this is import the whole sheet to a temporary table. Then, with a series of Delete queries, I remove the rows that have blanks where data MUST be in order to be valid. Also, where a field is not numeric ( NOT IsNumeric(x) ) but must be numeric to be valid. Once I've filtered out the trash, what is left is good data, which I then import to my final table.
Nobody said it would be easy, but really, it IS easy. Just tedious.