Credit Scores Explained: The Five Factors and What Actually Moves the Number

August 16, 2026 · guides · 11 min read

A credit score is a compressed prediction. It answers one narrow question — how likely is this borrower to fall 90 days delinquent on an account in the next 24 months — and compresses it into a three-digit number. It is not a measure of wealth, income, or financial competence, and it does not attempt to be.

Understanding what goes into it matters because the score is the single input that prices most household borrowing, and the difference between bands is measured in tens of thousands of dollars over a mortgage. This guide covers the five factors and their weights, the mechanics that trip people up, and what a band is actually worth.


The Five Factors and Their Weights

FICO publishes the approximate weighting of its general-purpose scores:

Factor Weight What it measures
Payment history 35% Whether accounts have been paid on time
Amounts owed (utilization) 30% Balances relative to available credit
Length of credit history 15% Age of accounts, average and oldest
Credit mix 10% Variety of account types
New credit 10% Recent inquiries and newly opened accounts

VantageScore, the competing model, uses different weights and terminology but keys off the same underlying data. Lenders also use industry-specific variants — an auto lender may pull a FICO Auto Score weighted toward auto loan behavior, a card issuer a FICO Bankcard Score. This is why the number a consumer sees in an app frequently differs from the number a lender quotes: different model, different version, sometimes a different bureau.

Two-thirds of the score sits in the first two factors. Almost everything worth doing lives there.


Payment History (35%)

The largest factor and the simplest: has each account been paid as agreed.

What registers:

Most negative marks remain on a report for seven years from the date of first delinquency; Chapter 7 bankruptcy remains for ten.

The important nuance: impact decays. A 30-day late from four years ago carries far less weight than one from four months ago. A single late payment on an otherwise clean file with long history typically produces a drop that recovers substantially within a year or two, even though the mark itself stays visible for seven.

Autopay for at least the minimum on every account is the single highest-leverage mechanical step available, because it eliminates the failure mode that carries the most weight.


Amounts Owed (30%) — and the Mechanic Most People Get Wrong

This is overwhelmingly credit utilization: revolving balances divided by revolving credit limits.

Two versions are calculated, and both matter:

A single card maxed out hurts even when aggregate utilization is low. Both figures are inputs.

The mechanic people get wrong: card issuers report to the bureaus on the statement closing date, not the payment due date. A household that charges $4,000 a month on a $10,000-limit card and pays the statement in full, every month, without ever carrying a balance or paying a cent of interest, is nonetheless reporting 40% utilization — because the balance snapshot is taken when the statement closes, before the payment posts.

Two ways to address it:

  1. Pay before the statement closes, not merely before the due date. Making a payment a few days ahead of the closing date reports a much lower balance.
  2. Request a limit increase. Utilization is a ratio; raising the denominator lowers it without changing behavior. Many issuers process limit increase requests with a soft inquiry.

Where the thresholds sit: score impact is roughly continuous rather than cliff-shaped, but the commonly cited inflection is around 30%, with the best outcomes generally in the 1–9% range. Reporting exactly 0% across every card is marginally worse than reporting a small positive balance on one — the model reads a completely dormant file as less informative than an actively and responsibly used one.

Utilization also has a property no other factor shares: it carries no memory. Pay balances down this month and next month's report reflects the lower figure with no residual penalty. It is the fastest-moving lever in the entire score.

Installment loan balances (mortgage, auto, student) are counted separately and weigh far less. Carrying a large mortgage does not damage a score the way a maxed-out card does.


Length of Credit History (15%)

Three sub-inputs: age of the oldest account, average age across all accounts, and how long since each account was last used.

The practical consequences:

Where an old card carries an annual fee, asking the issuer to product-change it to a no-fee card in the same family typically preserves the original account opening date. Closing and reapplying does not.


Credit Mix (10%) and New Credit (10%)

Credit mix rewards having both revolving accounts (cards) and installment accounts (loans). At 10% it is not worth taking on a loan to improve — the interest cost dwarfs the score benefit. It is worth knowing about mainly because paying off the last installment loan can produce a small, temporary dip, which surprises people who expected the opposite.

New credit covers hard inquiries and recently opened accounts. A hard inquiry typically costs a few points and fades within a year, dropping off the report entirely after two. The details that matter:


What Does Not Affect the Score

Common misconceptions, all false:

Lenders do consider income and employment — they are simply not inputs to the score itself. A mortgage underwriter evaluates debt-to-income ratio and score as separate criteria.


What a Band Is Actually Worth

The score matters because of what it prices. Approximate score bands:

Band Range
Exceptional 800–850
Very good 740–799
Good 670–739
Fair 580–669
Poor 300–579

The financial consequence is easiest to see on a mortgage, where lenders publish tiered pricing. A representative spread between the top tier and a 660–679 score is roughly 1.0 to 1.5 percentage points of interest rate.

On a $400,000 30-year fixed mortgage:

Rate Monthly principal and interest Total interest over 30 years
6.25% $2,463 $486,600
7.50% $2,797 $606,900
Difference $334/month ~$120,300

That is the practical case for treating the score as a maintenance item well ahead of a major borrowing event, rather than as something to address once an application is already in motion.

The spread compresses at the top. The difference between 760 and 820 is usually zero in pricing terms — most lenders' best tier begins somewhere in the 740–780 range. Optimizing a 780 score toward 830 has essentially no financial value; moving a 640 to 700 has a great deal.

Auto loans show a wider proportional spread than mortgages. Insurance premiums in most states are also priced partly off a credit-based insurance score, a related but distinct model.


Realistic Repair Timelines

What moves quickly and what does not:

Action Typical time to show
Paying down card balances 1–2 statement cycles (30–60 days)
Requesting a credit limit increase 1 statement cycle
Disputing and removing an inaccurate item 30–45 days
Hard inquiry impact fading 6–12 months
Recovering from a single 30-day late 12–24 months
Recovering from a collection or charge-off 2–4 years, then falls off at 7
Building average account age Years; no shortcut exists
Building a file from nothing 6 months minimum before a FICO score can be generated

The realistic ceiling on rapid improvement is what utilization alone can deliver — which, for someone carrying high balances, can be substantial. Someone at 85% utilization who pays down to 8% can see a large move in two months. Someone whose problem is a recent charge-off has no fast path.

Every consumer is entitled to free weekly reports from all three bureaus through AnnualCreditReport.com, the site authorized under federal law. Reviewing all three matters because they do not always contain the same information, and errors are common enough to be worth checking for before any major loan application.

A note on credit repair services: nothing a paid service can do is unavailable to a consumer directly. Disputing genuinely inaccurate information is free and effective. No legitimate service can remove accurate negative information, and any that claims otherwise is describing something that does not exist.


Frequently Asked Questions

Why do the three bureaus show different scores? Because creditors are not required to report to all three, and many report to only one or two. Different underlying data produces different scores from the same model. Differences of 20–40 points across bureaus are ordinary.

Does carrying a small balance help the score? No — this is one of the most persistent myths. Utilization is measured at the statement snapshot, so a card that reports a small balance and is then paid in full produces the same utilization figure as one where a balance is carried, with no interest cost. There is no scoring benefit to paying interest.

Does closing a paid-off card help? Generally the opposite. It removes that card's limit from the utilization denominator immediately, and eventually reduces average account age. Where an annual fee is the issue, a product change to a no-fee card usually preserves the history.

How long does it take to build a score from zero? FICO requires at least one account open for six months with activity reported in the last six months. A secured card or a credit-builder loan is the usual entry point. Reaching the 700s from a standing start typically takes one to two years of clean history.

Does becoming an authorized user help? It can, meaningfully, since the primary account's history may be imported onto the authorized user's file. The effect depends on the scoring model version and on the primary account's own behavior — an authorized user on a maxed-out, late-paying card inherits the damage as well.

Does a credit score affect a rental application or a job? Landlords commonly pull credit reports and sometimes scores. Employers in most states may pull a modified credit report with written consent, but not a score, and several states restrict the practice further.


This content is for educational and informational purposes only and does not constitute financial, credit, or legal advice. Equity Rank is not a registered investment adviser. Scoring model weights are approximate and published by the model vendors; specific lender pricing tiers vary. Rate examples are illustrative amortization calculations, not quotes.