Float optimization: the money in motion
Funds in flight are an asset class with their own optimisation problem.
By Solomon Ajayi · Free to read, no signup
Float is money that's left one account but hasn't arrived at the next, clearing checks, T+1 card settlements, ACH in transit, inbound wires being verified. From your books, float sits as IN-TRANSIT assets: not yet usable but contractually yours. Smart treasury teams MINIMISE outbound float (get money out of your accounts as late as possible while still meeting commitments) and MAXIMISE inbound float (get money INTO usable accounts as fast as possible). This lesson posts an inbound batch of card settlements and shows how same-day-vs-next-day settlement timing changes when interest income starts.
Money in motion is still money, but it is not yet earning. When a card batch captures on Day 0, the funds are contractually yours, sitting in a Provider Receivable, but they are not in a bank account where they accrue the overnight rate. That gap, between when value is owed to you and when it lands somewhere productive, is float. It looks invisible because nothing in the journal ever debits the yield you missed, yet it is a real cost paid every single day.
The settlement entry itself is unremarkable: Bank Account up, Provider Receivable down, a plain asset-to-asset move. The economics hide in the timing. At a 6% overnight rate, 24 hours of T+1 float on ₦50M is roughly ₦8,200 of foregone yield, and across a year of daily batches that compounds into millions. Same-day settlement closes the gap but charges a flat fee, so the choice is a daily arithmetic problem: does the recovered yield on today's batch size beat what the processor charges?
Because the cost of float never posts as a journal line, it is easy to pretend it does not exist. Treasury teams make it visible by quantifying it as a shadow number on a daily dashboard and budgeting against it, the same way you would track any other expense. At a hundred million naira a day in flows, optimising float moves the needle as much as negotiating fifty basis points more on your deposit rate.
Worked example, step by step
Day 0: ₦50M of card sales captured (T+1 settlement)
Daily card settlement batch. Standard T+1 means the provider holds the funds overnight and lands them tomorrow. From your perspective: provider receivable of ₦50M, not yet in your bank account, not yet earning treasury interest.
| Account | Debit | Credit |
|---|---|---|
| Provider Receivable (T+1) (1100) | ₦50,000,000.00 | |
| Same-Day Provider Receivable (1110) | ₦50,000,000.00 |
Provider Receivable (T+1) UP ₦50M (debit, asset). Originating revenue / wallet entries omitted to focus on the float position.
Day 1: settlement lands, money moves to bank
Next morning, ₦50M settles to your bank account. From now on (a full 24 hours late vs same-day settlement) the money earns the overnight rate. At a 6% APR, 24 hours of float cost you 50M × 6% × 1/365 ≈ ₦8,200. Across 365 days a year of T+1 batches that's ₦3M of foregone yield.
| Account | Debit | Credit |
|---|---|---|
| Bank Account (Overnight) (1610) | ₦50,000,000.00 | |
| Provider Receivable (T+1) (1100) | ₦50,000,000.00 |
Bank Account UP ₦50M (debit). Provider Receivable (T+1) DOWN ₦50M (credit). The 24 hours of foregone interest never appears as an entry, it's a SHADOW cost (you didn't earn what you could have). Treasury teams quantify this as 'cost of float' and budget against it.
Optional same-day settlement (paid for): cost ₦25,000
Your processor offers same-day settlement for a flat fee of ₦25,000 per batch. With ₦50M × 6% × 1/365 ≈ ₦8,200 of recoverable yield, the math doesn't work, you'd LOSE ₦16,800 paying for same-day. The break-even on this rate / processor combo is around ₦150M/batch. Below that, T+1 is correct.
| Account | Debit | Credit |
|---|---|---|
| Same-Day Provider Receivable (1110) | ₦0.01 | |
| Provider Receivable (T+1) (1100) | ₦0.01 |
If you'd chosen same-day: Same-Day Provider Receivable UP ₦50M debit immediately (vs T+1 receivable), Bank Account UP ₦50M debit minutes later, Fee Expense UP ₦25,000. The decision is RATE-DEPENDENT and BATCH-SIZE-DEPENDENT, your treasury job recomputes this every day given today's rate and forecast batch sizes.
Takeaway
Float is a real asset class with a real optimisation problem. Inbound float (money owed to you not yet received) costs you the foregone yield until it arrives, minimise it with same-day settlement IF the rate × batch-size justifies the fee. Outbound float (money you owe but haven't paid yet) is the OPPOSITE: hold onto it until the latest moment your SLA allows. At scale (₦100M+/day in flows), float optimisation moves the needle as much as a 50bps better deposit rate would. Build the recon and forecast tooling early; treat float as a first-class item on your daily treasury dashboard.
Practice this on a real ledger
Reading is half of it. Open this lesson in the lab to post the entries yourself against a real Postgres-backed double-entry ledger, with the validation on. Free, your sandbox is yours.