What actually moves a credit score, in order of weight
Five factors, and they are not equally weighted. Two of them account for most of the movement, which means most of the advice you have read is aimed at the wrong things.
Credit scoring attracts an unusual amount of folklore. Part of the reason is that the models are proprietary, so there is room for speculation. Part of it is that the weightings are public in broad terms and widely ignored.
The weightings are the useful part, because they tell you which actions are worth taking.
An important caveat first
There is no single global credit score.
In the United States, FICO and VantageScore dominate, and FICO publishes approximate category weightings. In the United Kingdom, each of the main credit reference agencies produces its own score on its own scale, and lenders largely use their own internal models. In India, CIBIL and other bureaus score on a 300–900 range. Other countries differ again, and some have no consumer score at all.
The weightings below reflect the widely published FICO breakdown, which is the most transparent model available. The direction of each factor holds up broadly across systems; the exact percentages do not travel. Check what applies where you live.
The five factors, by weight
Payment history — around 35%
Whether you have paid on time. The largest single factor by a clear margin, and the one with the longest memory.
A missed payment reported to a bureau can affect a file for years — commonly six or seven, depending on jurisdiction. Recency matters: a late payment last month weighs more heavily than one four years ago. Severity matters too, with 30, 60 and 90-day markers escalating, and defaults, collections and court judgments considerably worse.
The practical implication is blunt. Paying on time, every time, is not one tip among many. It is the majority of the score. Automating at least the minimum payment on every credit account removes most of the risk of an accidental miss.
Amounts owed — around 30%
Chiefly credit utilisation: how much of your available revolving credit you are using. Total balances divided by total limits, and also calculated per card.
Lower is better, throughout. The commonly repeated “keep it under 30%” is a rough guide rather than a threshold — there is no cliff edge, and someone at 8% generally scores better than someone at 28%.
Two mechanics worth knowing:
- Timing. Utilisation is normally measured from the balance reported on your statement date, not after you pay. Paying in full every month can still show high utilisation if you spend heavily before the statement cuts. Paying down before the statement date, rather than before the due date, is what moves this figure.
- Instalment debt counts differently. A mortgage or car loan sits in this category too, but the proportion of the original balance remaining is treated far more gently than revolving utilisation.
The two that matter most
- Never miss a payment — automate the minimum on everything
- Keep reported balances low relative to limits, and pay before the statement date
- Everything below accounts for roughly a third of the score combined
Length of credit history — around 15%
The age of your oldest account, the average age of all accounts, and how long each has been active. Older is better, and this factor moves slowly by nature.
The actionable implication is mostly negative: closing an old account can hurt twice over, by removing its age from the calculation eventually and by reducing your total available credit, which pushes utilisation up immediately. If an old card has no fee, leaving it open with occasional small use is usually the better option.
New credit — around 10%
Recent applications and newly opened accounts. Each formal application generates a hard search, which typically has a small effect and, in the FICO model, stops being counted after twelve months while remaining visible on the report for longer.
Several applications in a short period look worse than one, because the pattern suggests urgency. Rate-shopping for a single mortgage or car loan within a short window is often treated as one event by the newer models, but the rules vary — check before making multiple applications.
Checking your own score or report is a soft search and has no effect.
Credit mix — around 10%
Having both revolving accounts and instalment loans. The smallest factor, and not worth taking on borrowing you do not need in order to improve it.
What has no effect
Worth stating plainly, because these come up constantly.
- Your income, savings balance or employer. Lenders consider income in their own decisions, but it is not in the score.
- Checking your own credit report or score.
- Using a debit card, or paying in cash.
- Your age, marital status or where you were born.
- Being married to someone with a different score. Files are individual, though joint accounts appear on both.
Most of the advice circulating is aimed at the bottom three factors, which together account for about a third of the score. The top two account for around two thirds.
Where the leverage is
If you want the score to improve, in order:
- Fix any missed payments and never add another. Automate minimums.
- Reduce reported balances, particularly on cards near their limits, and pay down before the statement date.
- Stop closing old accounts unless there is a fee that justifies it.
- Space out applications rather than clustering them.
- Check your report for errors. Incorrect late markers, accounts that are not yours, or a balance that should be zero all happen, and all are correctable. This is the one step that can produce a fast improvement.
- Wait. Time genuinely does the rest, and no legitimate service can accelerate it.
On credit repair offers
Anything promising to remove accurate negative information for a fee is selling something it cannot deliver. Accurate records can be corrected only if they are wrong. Inaccurate records you can dispute yourself, free, directly with the bureau.
If you are dealing with problem debt rather than an optimisation question, a non-profit or government-backed debt advice service is the right place to start. Their advice is free, and it will be better than anything on a comparison page.
This article describes how scoring models generally work. It is not advice about your own file or about any credit product, and models differ by country and change over time. If a credit decision matters to you, check your report directly with the bureau and consider speaking to a qualified, regulated adviser.
