Utilisation: the 30% rule of thumb, and what it gets wrong

What utilisation actually is

Revolving utilisation is the balance reported on your revolving accounts divided by the limits reported on them. Both words matter. It is not what you spent, it is not what you owe today, and it is not what your banking app shows. It is a number computed from a snapshot a furnisher sent to a bureau on a particular day.

It sits inside “amounts owed”, which is roughly 30% of the FICO model — the second-largest factor, and by a wide margin the fastest one to move, because it has no memory. Unlike a late payment, which stays on the file for seven years, utilisation is recomputed every time a new balance is reported. A high figure last month has no residual effect once a lower one arrives.

The statement-date trap

Here is the thing that catches almost everybody. Most card issuers report your balance to the bureaus on or shortly after the statement closing date, not after the payment due date. The due date is typically three weeks later.

So if you charge $1,800 on a $2,000 card, let the statement close, and then pay it off in full before the due date, you have paid no interest — and you have also reported 90% utilisation. Someone who pays in full every single month, has never been late, and carries no debt at all can present as a maxed-out borrower on the file. We see this constantly, and the people it happens to are usually the most financially careful ones on our list.

The fix costs nothing: find out each card’s statement closing date (it is on the statement, and support will tell you in thirty seconds), and make a payment two or three days before that date rather than waiting for the due date. Pay the rest as usual. Nothing else changes.

Where the 30% rule of thumb came from, and what it gets wrong

The 30% figure is a rule of thumb that escaped into folklore. There is no threshold in the scoring model at 30%. The relationship is continuous: lower is better essentially all the way down, and people with the highest scores generally report utilisation in the low single digits rather than at 29%.

Two things the rule gets wrong:

• It implies 29% is a target. It is not a target; it is a ceiling that is better than 60% and worse than 6%.
• It implies 0% is optimal. It is not quite. Reporting zero across every revolving account can score slightly worse than reporting a small positive balance, because the model reads “no activity”. Leaving one small charge to report on one card, and paying the rest before the statement, is the usual advice.

Per card, and in aggregate

Utilisation is measured twice: across all your revolving accounts together, and on each card individually. A single card sitting at 96% can hurt even when your aggregate figure is comfortable, which is why moving a balance between cards is not always a neutral act.

This also explains a common piece of bad advice. “Consolidate everything onto one card” typically improves nothing and can hurt: it takes one account to a very high individual utilisation while leaving your aggregate unchanged.

A worked example, with the arithmetic shown

Four cards, $16,700 of total limit, and $6,400 of balance. In the first arrangement the balance is concentrated on the small-limit cards. In the second, the same total is arranged so the small cards report low and the largest limit carries most of it.

Aggregate utilisation is 38% in the first arrangement and 38% in the second — nearly identical, because the total balance has not changed. What has changed is that two cards have come off 92% and 96%. That is the part a per-card check would see and an aggregate check would miss.

The second arrangement is better on the per-card measure. It is not free: moving a balance may cost a transfer fee, and it does nothing at all about the underlying debt. The genuinely better move is the one in the paragraph above — pay before the statement closes — because it lowers the aggregate as well.

Three expensive mistakes

Closing a card you have paid off. The limit disappears from the denominator the day it closes, so your utilisation goes up immediately even though you did nothing. If the card has no annual fee, keep it open and put one small recurring charge on it.

Asking for a limit increase at the wrong moment. A higher limit lowers utilisation, which is helpful — but some issuers process the request as a hard inquiry. Ask whether the increase is a soft pull before you request it, and never request one in the ninety days before a mortgage application.

Believing a company that quotes you points. Utilisation is a real and substantial lever. It is still one input into a model nobody outside FICO has, applied to a file nobody has seen but you. Anyone who converts a utilisation change into a specific number of points is making it up.