The trouble with all of this is that it’s still creating a multi-mailpiece record, which means the inserter marks are going to be broken too, as they are only going to recognize each record as a mailpiece rather than the 2-3 mailpieces it actually is.
I honestly believe the only sane way to do this would be to pre-process the data to make the splits. Set it up such that one mailpiece equals one record so that the inserter marks can be generated on the page correctly without requiring a second pass through Connect.
Effectively, every method I can think of that doesn’t pre-process the data ends up requiring two passes through Connect in some way, which isn’t ideal. By pre-processing the data, we can end up in an ideal situation for handling the inserter marks in one pass through Connect.
Of course, depending on what sort of data we’re talking about, that pre-processing is going to be more or less difficult. CSV where the address information appears on each line, for example, could be done in the datamapper with nothing more than a little boundary script forcing a break every time the address changes or every time we’ve hit X records. Other data types would probably be more tricky.