The waiting inside
a cheaper payment.

Combining small charges can reduce the fee paid for each purchase. The saving needs a clock attached: when does the customer pay, when can the seller use the money, and which institution carries the gap?

A digital service sells small units of work. Rather than collect each charge immediately, it lets them accumulate. The eventual payment is larger, and a fixed processing charge is spread across more purchases. This can be a sensible design. It also changes the service being compared.

My published research on sub-dollar digital payments distinguishes fixed-plus-percentage fees from byte-denominated charges. Its 55,000 evaluations combine schedule-computed legacy fees with transaction-measured SPV fees. The comparison concerns direct processing charges, with other service costs outside its scope. A further design question follows: if aggregation reduces a fixed charge, what timing and funding arrangement makes that aggregation possible?

Put a clock beside the fee

Consider a deliberately simplified illustration. Each purchase has value v. Processing a payment costs a fixed amount a plus a fraction b of its value. Assume the service can legally and operationally combine N purchases into one charge, with no extra aggregation fee, and that the same schedule applies to the combined amount. The average direct fee per purchase becomes a/N + bv, rather than a + bv.

Use invented figures: a 10-cent purchase, a 20-cent fixed fee and a one-percent proportional rate. Separate collection costs 20.1 cents per purchase. Collecting ten purchases as a single one-dollar charge costs 21 cents in total, or 2.1 cents per purchase. These are arithmetic examples, not prices offered by a provider.

Now suppose purchases arrive exactly one minute apart and the service charges immediately when the tenth arrives. The first purchase waits nine minutes for collection; the last waits zero. Average waiting is 4.5 minutes. More generally, with spacing d, the mean wait within a completed batch is (N − 1)d/2. This calculation excludes processing time and assumes every batch fills.

Quiet periods change the problem. A count threshold may leave a final incomplete batch waiting. Adding a time limit controls that delay but can produce smaller batches and higher fees per purchase. An average calculated only from completed batches would conceal the unfinished work.

Follow availability and settlement separately

Batch charging in this example must be distinguished from netting obligations between payment institutions. The BIS analysis of fast payments explains that a payee can receive available funds before settlement between payment service providers. In a deferred arrangement, the receiving provider can advance the funds; this creates an exposure between providers that needs risk controls. Real-time settlement instead requires liquidity when individual payments settle.

Consequently, “the seller receives funds immediately” does not, by itself, identify the settlement model. Nor does deferred inter-provider settlement necessarily mean that the seller waits. The relevant evidence identifies both events and the party funding the interval. This is a distinction between arrangements, not a claim about the current design of any named service.

Keep the comparison at the same level

For the hypothetical digital service, I would compare immediate collection, delayed collection and an advance-funded arrangement against the same sales sequence. The record would retain each purchase time, collection time, funds-availability time, amount and fee. Any financing charge or unrecovered balance would be recorded separately rather than folded into an unexplained processing rate.

The evaluation would include a quiet period, a customer who stops buying before the threshold, and a failed collection. It would report outstanding balances and waiting-time tails alongside the mean. These are proposed tests, not results of a new field study.

A smaller direct fee can improve the economics of a small sale. Whether batching delivers the better service depends on what happens between the sale and usable payment. Connecting the cost schedule to those events makes the trade-off examinable: the fee saving, the waiting it introduces, and the funding arrangement that may remove that waiting from the seller’s view.