The debt snowball vs avalanche argument is usually settled by temperament. People who like spreadsheets pick avalanche, highest interest rate first, because it is provably the cheapest order. People who have abandoned a plan before pick snowball, smallest balance first, because closing an account is a feeling, and feelings are what keep you paying in month fourteen. Both camps are right about their own point. What neither side tends to say is how much money is on the table, so I measured it, first on one household and then on nineteen thousand.
The debt payoff calculator runs both orders side by side, and the same warning sits under every figure in this post. The engine assumes you add nothing new to any balance and that every rate holds for the whole run. Cards reprice, a 0% window ends on a date the projection does not know about, and a card still in use breaks the plan faster than either ordering can save it. Treat the gap it reports as the price of a choice, not a forecast of your interest bill.
What the debt snowball vs avalanche gap costs on one pile
The calculator opens with three debts: a credit card at £6,000 and 21.9%, a store card at £1,200 and 9.9%, and a car loan at £8,000 and 7.9%. The minimums come to £335 a month, the slider adds £150 on top, and the pile is £15,200. Those are the numbers already on your screen if you open the tool, so nothing below needs typing in.
Avalanche sends the £150 at the card. It clears in month 29, the store card follows in month 31 and the car loan goes in month 39, with £3,271 of interest paid along the way. Snowball sends the same £150 at the store card instead. That is gone in month 7, the credit card follows in month 31 and the car loan again in month 39. Interest: £3,547.
So the order is worth £276 on this pile, and not a single month. Spread over thirty-nine months that is about £7 a month, which is the cost of having a debt disappear in month seven rather than month twenty-nine.
Now move the slider instead. Push the extra from £150 to £250 and avalanche finishes in 31 months at £2,437, eight months and £834 better than it was. Drop the extra to nothing and both orders take 69 months and £7,822, because with minimums only there is no spare money to aim and the strategies become identical. Or keep snowball and add £25: at £175 extra it pays £3,263, below avalanche's £3,271 at £150, and finishes two months sooner. On these debts the whole ordering argument is worth £25 a month.
That is the usual shape, but it is not the only one. Make the credit card £9,000 at 29.9% with a £180 minimum and the picture flips, because the biggest balance is now also the dearest. Snowball clears the store card in month 7, then spends the next twenty-one months on the £8,000 car loan at 7.9% while the card compounds at 29.9% behind it. It finishes in month 56 with £11,825 of interest. Avalanche goes straight for the card and finishes in month 48 with £7,938. Eight months and £3,887, from the same money.
There is a middle path for that pile, and the engine can price it. Take the one quick win, the store card in month 7, then switch to avalanche with its £35 minimum rolled into the extra. That finishes in month 50 with about £8,800 of interest: the quick win costs roughly £870 and two months, not £3,887 and eight. You do not have to choose a camp. You can buy one early closure and then be sensible.
And sometimes there is nothing to decide. Put the store card at 39.9% and it is both the smallest balance and the highest rate, so both orders attack it first and the tool reports no difference at all. Give every debt the same APR and the orders differ but the interest does not, since money aimed at any balance earns the same rate wherever it lands.
Nineteen thousand piles
One household could be a fluke, so I generated 20,000 random ones and ran each through both orders: two to five debts, balances from £300 to £15,000, APRs anywhere from 0% to 40%, minimums between 1% and 4% of the balance with a £25 floor, and an extra payment between £50 and £500. In 757 of them the payments never outran the interest under either order, which is the calculator's stalled warning and a sign the problem is the minimums, not the sequence. The other 19,243 cleared.
In a quarter of those, 26%, the smallest balance was also the dearest, so the two strategies paid in the same order and the gap was zero. Across the whole set the median gap was £793, or 11% of snowball's interest bill. It stayed under £500 in 42% of piles and went above £1,000 in 46%, above £2,000 in 32%, and one pile in ten lost more than £5,400 to the order. The tail is long because the piles that punish snowball are the ones with a big balance at a card rate, and those exist.
Time is the quieter result. The median pile finished one month later under snowball. One in ten finished six or more months later, one in twenty ten or more, and no pile finished sooner. What snowball buys instead is the first closure: the median first debt cleared in month 11 under snowball and month 20 under avalanche, nine months sooner for the win people say they need.
The number I find most useful is the exchange rate between the two. For the first 3,000 piles I asked how much more extra payment snowball needed each month to match avalanche's interest bill. In 2,859 of them the answer was £600 or less, with a median of £60 a month and a mean of £99. Put plainly: for the typical pile, an order you will stick to plus £60 a month beats the optimal order on its own. Where the gap runs into thousands, it is because £60 would not cover it, and the tool's strategy check line tells you which kind of pile you have.
Whether the quick win keeps anyone paying is a separate question, and the evidence is thinner than the advice. Alexander Brown and Joanna Lahey tested the idea in a laboratory task for the National Bureau of Economic Research: when an unpleasant job was broken into parts of unequal size, people completed it faster with the parts arranged smallest to largest. They also found that, offered the choice of ordering, people picked smallest-first least often. A lab task is not a credit card, and the paper says as much. But it is consistent with the calculator's numbers: the win is real, modest, and worth paying a little for if it is the difference between finishing and not.
Reading your own strategy check
Enter your balances, APRs and minimums from the statements rather than from memory, because a store card you remember as 22% is often 39.9% and that single figure can move the gap from hundreds to thousands. Then look at the line under the chart. It states the actual cost of snowball for your debts.
If it says a few hundred pounds and no months, pick whichever order you will still be following in a year, and spend your attention on the extra payment, which does more work than the order ever will. If it says thousands, your dearest debt is probably also your biggest, and avalanche earns its place; if you still want a closure early, take the one small debt first and switch, as the £9,000 example did. If the tool says stalled, ordering is not the problem. Free advice from StepChange is the better first step, and getting it early beats getting it after a default.
Two things come before any of this. A small buffer, even one month of essentials sized with the emergency fund calculator, stops the next unexpected bill going straight back onto the card you are clearing. And an employer pension match beats any consumer rate, so paying the minimum on a 20% card to collect a 100% match is not a close call. Beyond those two, a card at 21.9% is the best guaranteed return available to you anywhere, which is why this question is worth more than the mortgage one. The walkthrough of the calculator covers entering the figures and reading the chart, and when the last balance reaches zero, the £485 a month that was servicing this pile becomes the input to the FIRE calculator.
Run your pile. Read the one line. Then stop arguing about the order and move the slider.