Once a retailer integrates with Flux, they start sending us their transaction information. This information isn't personally identifiable - think of it as a stream of item-level data and transaction amounts, or a pile of anonymous receipts: we can see what was purchased, but we've got no way of identifying who did the purchasing based on the retailer information alone.
In order for us to be able to identify the transaction as belonging to a Flux customer, we'll need to be receiving the 'other side' of the transaction from the bank. It's only if we are receiving both sides of the transaction that we're able to identify a single transaction and create a digital receipt.
If we're not able to match the retailer transaction information with a bank transaction, then our matching algorithm assumes the transaction does not belong to a Flux customer and simply moves on to try to match the next transaction.