TLDR: A card purchase made on 20 September can sit outside any statement until 2 October and outside any payment until 27 October. A spending total closed on the 30th is therefore always incomplete. The fix is to date spending at the transaction and run a fixed close a few days into the new month.
Emirates NBD’s Key Facts Statement for credit cards puts the interest-free period at “25 to 55 days from the Credit Card Transaction date”. Same document, one line above: “Payments are due within 25 days post statement date”. The February 2026 price guide repeats it as a flat rule in the General table, 25 days, payment due date from the statement date.
That thirty-day spread between two cards is the single biggest reason personal spending totals are wrong. Not overspending. Timing.
Why a September spending total is not final until late October
Say the card statement date is the 2nd. The billing cycle then runs from the 3rd of one month to the 2nd of the next. A purchase made on 20 September lands inside a cycle that does not close until 2 October, and the money does not leave the current account until the due date twenty-five days after that, on 27 October.
So on 30 September, when most people sit down and ask what September cost, that purchase exists in exactly one place: a card app showing an unbilled balance that nobody thinks to open. It is not on a statement. It has not been paid. It shows up nowhere in the current account.
The result is a September that looks cheap and an October that looks catastrophic, because October absorbs September’s card bill on top of its own cash spending. Two months of readings, both wrong, and a conclusion drawn from them about discipline that has nothing to do with discipline.
Installment plans stretch it further. The same price guide states that for installment plans, the first installment “will be due from the second card statement following the booking date”. Book a plan on 20 September and the first installment does not appear until the November statement. The obligation started in September. The books catch up two months later.
The cut-off date is the whole trick
At the trading company where I work, nobody closes a month by looking at the bank balance. The month closes on a cut-off date, and the rule is that a cost belongs to the period in which it was incurred, not the period in which it was paid. Goods received on 20 September are a September cost even though the supplier gets paid in November on sixty-day terms. Payment terms move cash. They do not move the cost.
Personal tracking almost always does the opposite. It treats the card payment as the expense, because the card payment is the line that shows up in the current account. That single choice guarantees a lag of one to two months on every card transaction, and it double-counts nothing while mis-dating everything.
A cost belongs to the month you incurred it, not the month the bank finally took the money.
The correction is small and boring, and it is why Masrofna treats a card payment as a transfer rather than an expense. Record the purchase on the date it happened, against the card. Record the payment as a movement between the current account and the card. Groceries bought on the 20th are September groceries. The 27 October payment is money changing pockets.
Do that and two questions separate cleanly: what a month cost, and what a month’s cash flow looked like. They have different answers, which is why one number can never serve both.
A twenty-minute personal month-end close
Three steps, run on a fixed day. I use the 5th, because by then every statement for the previous cycle has arrived and nothing else is pending.
One, set the cut-off. Everything dated on or before the last day of the month belongs to that month. Anything after it waits. No exceptions for convenience, because the exceptions are what corrupt the comparison between months.
Two, clear the suspense pile. Every close produces a handful of lines nobody recognises, usually an acquirer name instead of a shop name. Categorise them in one pass rather than one at a time across the month. If a line is genuinely unidentifiable, park it in a single holding category and move on. In accounting that pile has a name, a suspense account, and the discipline is that it must be empty before the next close.
Three, take the closing balances. Current account, each card’s outstanding balance, cash on hand. Write them down. Next month’s opening balances are this month’s closing balances, and if they do not tie, the difference is a transaction that was missed, not a mystery.
Twenty minutes, once a month, and the resulting series is comparable month to month. That is the only property that makes a spending number useful. An accurate number for one month tells you nothing. Twelve consistent numbers tell you everything.
The close has a real deadline, and it is not the budget
There is a harder reason not to let months pile up. The Central Bank’s Consumer Protection Standards require, at clause 5.1.1.46, that consumers be allowed a minimum of 30 business days to report an unauthorised transaction to their bank after being properly informed of it. Properly informed usually means the statement. A close run three months late puts the window behind you, and a legitimate claim becomes a conversation about goodwill.
Reconstructing that far back also costs money. The February 2026 price guide charges AED 15.75 for a duplicate statement covering the last three months, and AED 47.25 for anything older. The bank has priced staleness, and the price steps up at exactly three months.
None of this needs a fourteen-category system. It needs a date, a rule about which side of the date a transaction falls on, and the willingness to write down three balances once a month. Most tracking fails because it tries to be a real-time feed of an account that reports on its own schedule. Books do not work that way. Neither should a budget.