I've used an SQL Database GET connector at the start of the process to obtain all the previous days transactions, and then use this information to find appropriate bank accounts in NetSuite according to the currency used on the transaction, as well as using the ID of the transaction data as a parameter to look for transactions in NetSuite.
The only issue is, I need to run each unique ID from the SQL DB individually, and as we know we can't use a Data Process shape to split database profiles. I could use the Grouping Options in the DB connector to use each ID as a link element, but this would mean that for every single unique ID (there will be 1000-2000 per day), it would run a query for accounts in Netsuite, and also Transactions in NetSuite, which isn't very efficient at all.
It doesn't work to create another SQL DB connector off the START shape like so, as it still treats it as a single batch of transactions, and I end up with transactions entering the same lines into the transaction UPSERT
Does anyone know of a way that I can efficiently use the data obtained in the START shape to still query the accounts and transactions within Netsuite, but then use the same data set obtained to use a link element of the ID with a batching of 1 to then go through the rest of the process. Would I need to run a sub-process perhaps?