An Insert process can always insert all the data successfully regardless of the presence of complex referential integrity (RI) cycles.

An Insert process allows for Update/Insert processing in one step; a Load process may not.

An Insert process will be faster than a Load process when inserting large volumes of data.

An Insert process allows you to use Column Maps to transform data; a Load process does not.

Optim will not allow you to create duplicate relationships between tables.

Optim will view two relationships between tables as possible duplicates and ignore one.

Optim will view two relationships between tables as possible duplicates but will still use both if selected.

Optim allows duplicate relationships but will ignore them if they are in fact duplicate.

Optim will keep cycling until all relevant rows have been found.

Reduced maintenance costs because application and database maintenance and license fees are no longer paid for the old application and its database, but data remains accessible to the user community.

Decommissioned data is needed every day by end-users and Optim caches it in memory so access is faster then the production database.

Reporting from the application against the archived data can be cheaper using tiered storage.

Optim can store decommissioned data in a compressed format so it will take less storage then it did in production.

Applications must continue to use data in a seamless manner.