Michael, That webcast was so helpful. It showed me I was doing it all wrong and went this route and it worked the first time!
the Key is this: The "repeat" is put around the the top section only because the API is building a CSV table to hold the read data. Then the <clist> section merely writes all the data at once. That was confusing until the host explained that.
Secondly, when the record ID is the one of the fields in the destination clist, it updates and if the record ID isn't there it adds. Very nice!
<qdbapi>
<usertoken>bxxxxxxxxxxxxxx</usertoken>
<records_csv>
<![CDATA[
%RepeatOn%
[new.SOP+SKU.csv],[new.Order Status.csv] <=== FIELDS you want the data from
%Repeatoff%
]]></records_csv>
<clist>6.7</clist> <=== FIELDS DATA IS GOING TO IN THE TABLE ABOVE (LINK)
</qdbapi>
