A credit score is a three-digit number between 300 and 850 that estimates how likely a borrower is to repay debt on time. FICO — the model used by 90% of top lenders according to FICO’s published data — calculates that number using five factors, with payment history (35%) and credit utilisation (30%) controlling nearly two-thirds of the result. A score of 670 or above is generally considered good; 740 and above unlocks preferred interest rates and terms from most lenders.

• Problem: Most people know they have a credit score but don’t know what’s actually in it, why it changes, or why the number looks different on different platforms.
• Solution: FICO scores use five factors — payment history (35%) and credit utilisation (30%) together control 65% of the calculation. Understanding these two factors alone makes credit management significantly less complicated.
• Result: Consumers with a 100% on-time payment record and utilisation below 10% consistently appear in the 800+ tier, according to FICO’s published methodology. Both are habits, not fixed characteristics.
• Source: FICO Published Scoring Methodology, VantageScore Technical Documentation, CFPB Consumer Credit Guidance, FCRA Statutes
• Time To Read: 9 Minutes

A three-digit credit score can affect access to mortgages, auto loans, credit cards, rental housing, and in some states, even insurance premiums. But the number itself is rarely straightforward.

I went through FICO’s published scoring methodology, VantageScore’s technical documentation, CFPB guidance on credit reporting, and the Fair Credit Reporting Act to put this guide together. What I found is that the system is more logical than most people realise — once you understand what the formulas are actually measuring.

Here’s how credit scores actually work in 2026.

Credit Score Ranges: What Is a Good Credit Score in 2026?

Most consumer credit-scoring models in the US operate on a 300 to 850 scale. A higher number signals lower historical default risk to lenders. What I found when reading through FICO’s published tier definitions is that the practical difference between score bands isn’t just a label — it directly determines which interest rates and products a borrower qualifies for.

Source: FICO published score tier definitions; CFPB credit score ranges guidance

One thing worth noting from the CFPB research: these bands are educational benchmarks — not statutory lending rules. A 760 score doesn’t guarantee loan approval, and a 640 doesn’t mean automatic denial. Lenders evaluate complete profiles including verified income, employment stability, and debt-to-income ratio alongside the score itself.

How Is a Credit Score Calculated? The 5 FICO Factors

When I worked through FICO’s publicly available methodology documentation, the thing that stood out most was how concentrated the formula is. Two factors — payment history and credit utilisation — control nearly two-thirds of the entire calculation. Everything else matters, but the math is heavily front-loaded toward these two habits.

Source: FICO published scoring factor weights

1. Payment History — 35%
   • This is the single largest factor in the FICO formula — by a wide margin. What the data shows is that FICO evaluates not just whether payments were missed, but how recently and how severely. A 30-day delinquency from two months ago creates a substantially larger score drop than a 30-day delinquency from five years ago. Severe items like bankruptcies remain on credit reports for up to 7 to 10 years under the Fair Credit Reporting Act.
   • The practical implication from CFPB research: setting up automatic minimum payments on all active accounts prevents accidental 30-day delinquencies — which carry disproportionate score damage relative to the amount missed.
2. Amounts Owed / Credit Utilisation — 30%
   • This factor surprised me when I first went through the FICO documentation, because it’s not simply about how much debt someone carries — it’s specifically about revolving credit utilisation. The calculation divides total credit card balances by total available credit limits.
   • For example: a $900 balance on a credit card with a $3,000 limit equals 30% utilisation. Paying that balance down to $300 drops utilisation to 10%. According to FICO’s published guidance, consumers in the exceptional 800+ tier consistently maintain overall utilisation below 10%. The 30% threshold is the commonly cited educational benchmark for preventing significant score deductions, but lower is consistently better in the data.
   • One important detail from the research: balances are typically reported to bureaus on the monthly statement closing date — not the payment due date. Paying down a balance before the statement closes means the lower balance is what appears in the scoring calculation for that month.
3. Length of Credit History — 15%
   • FICO’s methodology tracks three data points here: the age of the oldest active account, the age of the most recently opened account, and the Average Age of Accounts (AAoA) across the entire credit file. A longer track record gives scoring algorithms more data to evaluate financial reliability over time.
   • The practical implication: closing old credit cards — particularly no-annual-fee cards — can gradually reduce average account age once closed accounts eventually fall off credit reports. CFPB guidance on this point is consistent: keeping old no-fee cards open and occasionally active generally supports this scoring factor over time.
4. New Credit — 10%
   • This category tracks hard inquiries from formal credit applications over the preceding 12 months. What I found in FICO’s documentation is that it’s a relatively small factor — but it matters when multiple applications happen in a compressed timeframe, which signals potential cash-flow pressure to automated underwriting systems.
   • A formal inquiry is distinct from an opened account. The inquiry records the application; the new account separately affects average credit age and available credit. Both are tracked separately.
5. Credit Mix — 10%
   • FICO values experience managing different types of credit over time — both revolving accounts like credit cards and installment accounts like auto loans, student loans, and mortgages. This is the smallest factor in the formula and the CFPB research on this is clear: consumers don’t need to open unnecessary debt or pay interest purely to improve this category. It’s a supplementary factor, not a primary driver.

SaveXpert Credit Score Estimator

When I was researching how different actions affect credit score tiers, I found that most people significantly underestimate how quickly utilisation changes can move a score. The SaveXpert Credit Score Estimator lets you model your current inputs and see which tier your profile falls into:

Educational note: The estimator provides educational projections based on user inputs. It does not calculate official FICO Scores or guarantee lender approval decisions.

FICO vs. VantageScore 4.0: Key Differences

Here’s something that confused me when I first started researching this: the number on a free credit monitoring app and the number a mortgage lender sees can differ by 20, 30, or even 50 points — and both are technically correct. They’re just measuring the same credit file with different formulas.

FICO and VantageScore are competing companies with distinct mathematical algorithms. Here’s what the technical documentation shows about how they differ:

Source: FICO published methodology; VantageScore 4.0 technical documentation

What VantageScore 4.0’s Trended Data Actually Means

VantageScore 4.0 reviews up to 24 months of historical balance trajectory — not just a single monthly snapshot. What this means in practice is that the algorithm can distinguish between a “transactor” (someone who pays their full balance monthly) and a “revolver” (someone who carries a balance and pays minimums). That behavioural distinction is built directly into the score calculation in a way that the standard FICO model doesn’t capture.

Why Your Score Looks Different On Different Platforms

Based on what the technical documentation shows, there are five structural reasons why credit score numbers vary:

1. Different scoring models 
   • Free apps commonly display VantageScore 3.0 or 4.0, while FICO data shows 90% of top lenders use FICO models (FICO 8, FICO Auto Score, or FICO mortgage versions).
2. Different bureau files
   • Creditors may report monthly account data to only one or two of the three nationwide bureaus.
3. Reporting timing 
   • Lenders send updated balances to credit bureaus on different days, creating timing gaps across files.
4. Calculation timestamp 
   • A score pulled today reflects newer balance changes than one pulled two weeks ago.
5. Industry-specific versions
   • Mortgage lenders typically use FICO 2, 4, or 5; auto lenders use FICO Auto Scores scaled up to 900.

Hard vs. Soft Inquiries: What Actually Affects Your Score?

This is one of the most commonly misunderstood parts of credit scoring. When I went through the CFPB’s consumer guidance on inquiries, the key finding was straightforward: soft inquiries have zero mathematical impact on credit scores. The confusion tends to come from not knowing which type of inquiry a given action triggers.

The Rate-Shopping Protection Window

What the FICO documentation specifically shows is that the scoring model accounts for rate shopping. Multiple hard inquiries for the same loan type — mortgage, auto loan, or student loan — made within a specific window are grouped into a single scoring event:

• Newer FICO models and VantageScore: 45-day rate-shopping window
• Older FICO models: 14-day rate-shopping window

This consolidation does not apply to credit cards — every formal credit card application generates an independent hard inquiry regardless of timing.

Hard inquiries affect FICO scores for 12 months and remain visible on credit reports for 24 months, after which they’re automatically removed.

How to Check Your Credit Report for Free in 2026

One of the clearest things I found in the FCRA statutes is the distinction between a credit report and a credit score — and it matters practically. A credit report is the raw underlying record of all account data. A credit score is the mathematical calculation applied to that data. You need the report to verify what the score is actually being calculated from.

Under the Fair Credit Reporting Act, consumers have the legal right to inspect their underlying bureau files. Here’s what the law provides in 2026:

Your FCRA Dispute Rights

When incorrect balances, unauthorised accounts, or inaccurate late marks appear on a credit file, the FCRA provides a clear resolution process. According to the CFPB’s dispute guidance:

1. File a dispute directly with the bureau — Equifax, Experian, or TransUnion — online or by certified mail.
2. Notify the data furnisher — the creditor or lender that supplied the incorrect data.
3. Bureau investigates within 30 days — extendable to 45 days in certain circumstances.
4. Resolution notice — if the information cannot be verified, the bureau must delete or correct it and notify the consumer within 5 business days of completion.

One important legal point that the CFPB research confirms: the dispute process corrects inaccurate information — it does not remove accurate, verified negative history before the statutory period expires. Late payments stay on reports for 7 years. Chapter 7 bankruptcies stay for 10 years.

How to Improve Your Credit Score Step by Step

What the FICO methodology makes clear is that credit scores are not fixed characteristics — they’re calculated estimates that change as reported data changes. The five factor categories give a direct map of where to focus. Based on what the research consistently shows, here are the steps ordered by the size of their potential impact:

1. Protect Payment History (35% of Score)
   • CFPB consumer research shows that a single 30-day late payment can drop a score significantly — far more than the dollar amount of the missed payment would suggest. Setting up automatic minimum payments on all active accounts prevents accidental delinquencies. Even if the full balance isn’t paid, the on-time minimum payment preserves the payment history record.
2. Reduce Credit Card Utilisation Below 30% — Ideally Below 10% (30% of Score)
   • This is the fastest-moving lever in the formula. Because utilisation has no historical memory under standard FICO models — only the current reported balance matters — paying down a card balance can produce a score improvement as soon as the creditor reports the new balance to the bureaus, typically within 30 days.
   • Paying balances before the statement closing date (not just the payment due date) means the lower balance is what gets reported to the bureaus for that month’s calculation.
3. Keep Old No-Fee Cards Open (15% of Score)
   • Closing a credit card doesn’t immediately remove it from the credit report — it stays visible for roughly 10 years for accounts in good standing. But once it eventually drops off, the average account age decreases. CFPB guidance on this is consistent: keeping the oldest no-annual-fee cards open and occasionally active on small recurring charges supports this factor over time without costing anything.
4. Audit Credit Reports Regularly
   • An FTC study cited by the CFPB found that one in five consumers has an error on at least one credit report that could affect their score. Downloading reports weekly from AnnualCreditReport.com surfaces incorrect balance amounts, identity errors, or outdated collection notices before they cause ongoing score damage.
5. Structure Loan Applications Within Rate-Shopping Windows (10% of Score)
   • Using lender pre-qualification tools that use soft inquiries allows rate comparisons without any score impact. When formal applications are ready, executing mortgage or auto loan shopping within the 14-to-45-day rate-shopping window means multiple hard inquiries count as a single scoring event.

The Bottom Line

A credit score is a calculated estimate of lending risk based on reported account data — not a permanent judgment. The number changes as the underlying data changes, which is what makes the system manageable once you understand what it’s actually tracking.

What the research shows most clearly is that two habits drive the majority of the formula:

• Paying on time, every time — this single factor controls 35% of the FICO calculation
• Keeping credit card balances low relative to limits — this controls another 30%

Everything else — credit age, mix, and inquiries — matters at the margins. But the 65% that comes from those two habits is entirely within most people’s control regardless of where they’re starting from.

Free credit reports are available every week from all three major bureaus at AnnualCreditReport.com under federal law. Verifying what’s actually being reported — and disputing anything that’s wrong — is the foundation everything else builds on.