Why a four-payment split stops working on a large purchase

The four-payment split is a good mechanism for a narrow case. Two assumptions hold it up. The first is that the total is small enough that a quarter of it is not a decision — nobody checks a balance before agreeing to it. The second is that the whole arrangement is over quickly, inside a few weeks, so no one has to form a view about what a customer's finances will look like later in the year.

Both assumptions are true of a pair of shoes, a phone case, a weekly shop. They are true of most of what the model was designed around: small baskets, high volume, a checkout decision made in seconds and forgotten by the following month. Nothing about that is dishonest. It is a mechanism matched to a size. The problem is what happens when the size changes and the mechanism does not.

Here is the arithmetic, stated plainly. A large total divided four ways produces four instalments, each one a quarter of a large total. The model then compresses those four into the same short window it uses for a small basket — payments a fortnight apart, the last one due before the customer has finished the month after the purchase.

Consider what that asks. The person who could not pay the whole amount this month is now asked to pay a quarter of it this month, and another quarter before the month is out. If the number was large enough to stop someone at the counter, a quarter of it is usually still large enough to stop them at the counter. The affordability question has been rescheduled. It has not been answered.

The horizon is what actually breaks, more than the division. A short split does not reach far enough. The problem with a large purchase is that all of it lands inside one month's income, beside everything that month already wanted. Reaching past that means a term measured in months, not weeks. The four-payment model has no way to do it, because reaching further would mean underwriting further, and it was built specifically to avoid that.

The purchases where this matters are recognisable. Dental and elective procedures. School and university fees. A car repair, a set of tyres, a used purchase. A kitchen, a cooling system, a fit-out. A course of clinical care. These are not impulse buys, and the failure of a short split on them is not a failure of self-control. They are the purchases people defer for months and then make in one transaction, because the bill arrives whole and there is no version of it that arrives in parts.

They share one more property. The decision is usually already made before the price becomes an obstacle. Somewhere between deciding and paying, the question stops being whether and becomes when. That is the point at which the payment structure either helps or gets in the way.

When a short split does not fit, people do not go home and forget about it. They take one of two other answers, and both are worse than the thing they replaced.

The first is revolving credit. A card absorbs any number, which is precisely the problem: nothing in its structure imposes an end. The balance rolls, the minimum payment does not clear it, and the purchase is still being repaid long after anyone has stopped thinking about it. The debt outlives the reason for it, which is a different and worse condition than owing a known amount on known dates.

The second is postponement. The purchase is not cancelled, it is moved, and a decision already made does not get cheaper by waiting. The repair happens later, after the failure it was meant to prevent. The course starts a term late. The treatment goes ahead anyway, at the point where it has stopped being elective. Deferral looks free because nothing is charged for it. The cost lands somewhere else.

Longer is not automatically better, and it would be easy to read the argument above as a case for length on its own. It is not. A longer term is a longer exposure to whoever is on the other side of it, so the standard a plan has to meet rises with its term rather than relaxing.

Three properties decide whether a longer plan is trustworthy, and none of them is a rate. Every date and every amount visible before agreement, not disclosed afterwards in a statement. No revolving balance — a schedule with a defined end is a different instrument from a balance that renews itself, and the two should not be described in the same language. And fixed dates that do not move, because a schedule that the provider can adjust is not a schedule.

None of the three is a marketing claim. Each is a property of an agreement, which is why the agreement is the place to verify it. The terms and conditions are the authoritative text for financing through Cashew, and reading the schedule and the disclosure section of any provider's agreement before accepting it is the whole of the diligence available to a customer.

A merchant meets the same break from the opposite side. Someone has decided on the purchase, is standing at the counter, and the number stops them. A short split does not recover that sale, for the reason already given: a quarter of a large number is still a large number. The sale does not shrink to fit the payment option. It leaves.

An instalment plan long enough to fit a large basket settles differently for the seller as well. The merchant is paid the full amount on the sale and the provider collects from the customer over the following months, so the seller carries no collection risk and chases no one.

That arrangement is what the merchant side of this site is for. The merchant early-access form asks nine questions about the business and what it sells, and a person reads every answer and replies.

Treatment is the clearest case in the whole argument, for one reason: it is the large sum least able to wait. A quote for implants, a fertility cycle, a hair restoration or laser correction arrives as one number. The clinic needs paying at the point of care. And the decision has usually been made before the price is discussed, because the alternative to treatment is not a cheaper treatment.

Divide that quote into four and nothing useful happens. Each payment is a quarter of a clinical bill, due within weeks of a procedure the patient may still be recovering from. Place the same quote on monthly dates and the number changes character: the instalment becomes something that sits beside a salary, and the clinic is not asked to wait for its money while that happens.

The case is specific enough that a separate product exists for it. Hazel, Cashew's healthcare financing brand works in the UAE only and covers six categories of treatment: dental, fertility, cosmetic, vision, hair and scalp, and medspa. A patient applies at a partner clinic and spreads the cost of the treatment from there. The clinic is paid the full amount up front. It is operated by Cashew Technology Software L.L.C., the same company behind Cashew.

Everything that decides an application is published by Hazel rather than restated here. Hazel's eligibility criteria and terms are the authoritative version, and the one to read before applying.

One test is enough, and it can be applied at the counter. Take the total, divide it by four, and ask whether that number on its own would have needed thinking about. If it would, the split is the wrong instrument, and agreeing to it means agreeing to make the same difficult decision four times in quick succession.

The purpose of an instalment plan is not to make a large number look small. It is to put it somewhere it can actually be paid, on dates a person can see before they agree to them. That is a different design problem from the one the four-payment split was built to solve, and it needs a different mechanism.