Interest is taken first
Everything about credit card debt follows from one rule: the interest is charged before any of your payment reaches the balance.
interest this month = balance × (APR ÷ 12)
reduction in balance = payment − interest
On £5,000 at 22% APR, the first month's interest is about £91.67. Pay £150 and only £58.33 comes off the debt. Pay £100 and it is £8.33 — you have paid a hundred pounds and reduced what you owe by eight.
This is why the relationship between payment size and payoff time is not proportional. Doubling the payment does not halve the time; it can cut it by three quarters or more, because the increase lands entirely on the principal rather than being split with the interest.
It is also why there is a payment below which the debt never clears at all. If your payment is less than the interest charged, the balance grows every month regardless of how long you keep paying. The calculator says so outright rather than returning a number, because a figure like "412 months" implies progress that is not happening.
The minimum payment is the trap
Minimum payments look like guidance. They are not. They are the least the issuer will accept without treating the account as delinquent, and the formula makes that explicit.
A typical rule is 1% of the balance plus that month's interest, subject to a small floor. Read it carefully: the interest portion is exactly what it costs to stand still, so the only part reducing the debt is the 1%.
Worse, it shrinks. As the balance falls, 1% of it falls too, so the payment declines month after month and the tail stretches out. On a substantial balance at a typical rate this runs to decades, and the interest paid can approach or exceed the original amount.
The comparison row in the calculator exists for this reason. Set a realistic payment and look at what the minimum-only line says beside it — the gap is usually the most persuasive number on the page.
Regulators reached the same conclusion. US law now requires every card statement to show how long the balance would take to clear at the minimum, alongside the payment that would clear it in three years, precisely because the minimum was so widely misread as a recommendation. In the UK, firms must intervene when a customer has paid more in interest and charges than principal over eighteen months — the "persistent debt" rules.
Why a small increase does so much
The extra-payment field is worth using, because the effect is larger than intuition suggests.
Every additional pound goes entirely to principal. It is not shared with the interest, because the interest was already covered by the base payment. And each pound of principal removed stops accruing interest for the whole remaining life of the debt — so an extra payment early is worth far more than the same payment later.
On the default figures, adding £50 a month to a £150 payment on £5,000 at 22% cuts years off the term and saves a substantial multiple of the extra paid in.
The general shape: increases have their biggest effect when the payment is close to the interest charge, because a large share of what you were paying was being consumed. Once the payment is well above the interest, further increases behave more linearly.
How cards actually calculate interest
The calculator uses a monthly rate of APR ÷ 12. Real cards almost always do something slightly different, and the difference is worth understanding even though it is small.
Most issuers use a daily periodic rate: the APR divided by 365, applied to the balance each day, with the total charged monthly. Because each day's interest joins the balance the next day, it compounds daily rather than monthly, which costs marginally more.
At 22% APR the effective annual rate under daily compounding is about 24.6%, against 24.4% under monthly. In interest terms across a payoff, the gap is under one percent of the total — real, but not the thing that decides anything. The monthly approximation is the standard simplification and is used here for the same reason regulators use it in disclosure examples: it is close enough and far easier to follow.
Two related mechanics that matter more than the compounding difference:
The grace period. If you pay the statement balance in full each month, purchases typically carry no interest at all. This entire page is about balances that are carried; a card cleared in full every month is a free short-term loan and behaves nothing like the above.
Cash advances have no grace period. Interest starts on the day of the transaction, usually at a higher rate, and payments are generally applied to the highest-rate balance last under many older arrangements. A cash advance on a card being paid down is expensive in a way the headline APR does not show.
Several cards: avalanche or snowball
With more than one balance, the ordering question comes up.
Avalanche pays the minimum on everything and puts every spare pound at the highest interest rate first. It is mathematically optimal. It always costs less and always finishes sooner. There is no case in which it is beaten on arithmetic.
Snowball targets the smallest balance first regardless of rate. It costs more, and it produces a cleared account sooner, which is a real psychological event — there is reasonable evidence that people are more likely to stay with the plan.
The honest answer is that it depends on the spread. If your rates are close together, avalanche's advantage is small and the method you will actually finish is the better one. If one card is dramatically more expensive than the others, the maths gets large enough to override the motivation argument, and paying that one first is worth the loss of early wins.
Whichever you pick, the mechanism is the same: minimums on everything, everything spare on one target, and roll the freed-up payment onto the next as each clears.
Balance transfers, honestly
A 0% balance transfer can save a great deal. It can also quietly make things worse, and the difference comes down to two numbers people skip.
The fee. Typically 3% to 5% of the transferred balance, charged up front and added to the debt. On £5,000 that is £150 to £250 before you have saved anything. Compare it against the interest you would otherwise pay over the promotional period — usually it wins comfortably, but it is not free and it should be in the calculation.
The payment needed to finish in time. Divide the balance plus the fee by the number of interest-free months. That is the payment that clears the debt before the promotion ends. If you cannot make it, work out what you will still owe when the go-to rate begins — which is frequently higher than the card you transferred away from.
The failure mode is treating a transfer as making the debt cheaper to carry rather than faster to clear. The freed-up cash goes elsewhere, the balance is largely still there when the rate reverts, and the fee has been paid for nothing.
Also worth knowing: transfers usually do not earn the grace period on new purchases, so spending on a transfer card can start accruing interest immediately even while the transferred balance sits at 0%.
Where this sits against everything else
The recurring question is whether to clear the card or put the money somewhere else. At typical card rates the answer is unusually clear.
Paying off a balance at 22% is a guaranteed 22% return, tax-free, with no market risk and no time horizon. Nothing on the investing side offers that reliably. The usual comparison — long-run equity returns of perhaps 7% real — is not close.
The one common exception is an employer pension match, which is an immediate 50% or 100% return on the matched portion. Take the match, then attack the card.
An emergency fund is the other genuine competing claim, and it is not purely financial: without a small buffer, the next unexpected expense goes back on the card and undoes the progress. A modest fund alongside aggressive repayment is usually better than either alone.
Nothing here is financial advice. These are the mechanics of how the arithmetic works, and what it costs is a decision that depends on circumstances a calculator cannot see.