Phil Brammer
Edit: Sorry, this is a response based on SSIS. I'm leaving it here as it might help someone using SSIS. If you are seeking DTS advice, please seek the DTS group: http://groups.google.com/group/microsoft.public.sqlserver.dts lnk=srg&hl=en
This isn't the most elegant solution, but it will work and can be done without programming.
Since the file exists, it just might not have any records in it, you'll be creating two data flows. The first data flow connects to the file and hooks to a Row Count Transformation. That's it. Store the results of the Row Count Transformation into a package-scoped variable.
The second data flow is what you've already got built, just use it as is.
Connect the two data flows with a precedence constraint and use the expression "@Your_Variable > 0" to execute the second data flow if there were any records coming out of the Excel file.