Vladan,

Thanks for taking the time to create a different approach. I will research the "Number" table solution, as I am finding that I am using something similar in my latest scripts.

I also want to point out something very clever in your loading script:

One reason I was using Excel to generate my set of test records was that I could create a formula to preface the 000's on the keys. I will be using the statement you have above from here on out to do the same thing!

You noted that there is no mention of how to apply the new keys. You address this in your summary of inserting into a new table and using the new employee id from table two, which is a good solution. In my case, I was taking the new unused ids and using them in a conversion program written in Unibasic on a Unidata database. The Unidata tables were the source tables, and then DTS was bringing a copy of the tables to SQL Server, so the data had to be corrected in the underlying Unidata tables, but it was easier to identify the unused keys in SQL Server.