What a credit score represents
A credit score is a numerical summary of credit report data, designed to predict the likelihood of a borrower becoming seriously delinquent (90+ days late) within the next 24 months. It’s a risk assessment tool that compresses years of credit history into a three-digit number.
Lenders use scores because they’re efficient. Rather than manually reviewing every applicant’s complete credit history, a score provides a standardized risk metric that correlates with default probability across millions of borrowers. Higher scores indicate lower risk. Lower scores indicate higher risk.
The scores don’t measure financial responsibility or moral character. They measure patterns in data that statistically correlate with repayment behavior. Someone with a high income and no debt might have a lower score than someone with moderate income and lots of managed debt, because the scoring models evaluate credit management, not net worth.
The two major scoring systems
FICO scores dominate the lending industry, used in approximately 90% of lending decisions. FICO (Fair Isaac Corporation) has produced credit scores since 1989 and continuously updates its models. Current versions include FICO Score 8 (most widely used), FICO Score 9, and FICO 10. Each version weights factors slightly differently.
VantageScore was created by the three credit bureaus (Equifax, Experian, TransUnion) as a competitor to FICO. VantageScore 3.0 and 4.0 are the current versions. While gaining adoption, VantageScore remains less common in actual lending decisions, though it’s often what free credit score services show.
Both systems use the same underlying credit report data but weight factors differently. A person might have a FICO 8 score of 740 and a VantageScore 3.0 of 755 based on identical data. Neither number is “wrong” — they’re different models producing different outputs from the same inputs.
Additionally, each of the three credit bureaus maintains separate files. Not all creditors report to all three bureaus, so reports can contain different information. This means someone has at least six different scores at any time: FICO and VantageScore from each of three bureaus.
The five FICO factors
FICO scores are calculated from five categories of data, each with approximate weightings:
Payment history (35%): The record of on-time versus late payments across all accounts. A single 30-day late payment can drop a score significantly, especially if the score was previously high. More recent late payments hurt more than older ones. Severity matters: 90-day delinquencies impact more than 30-day. Collections, bankruptcies, and foreclosures are extreme negatives that remain on reports for 7-10 years.
Amounts owed (30%): Primarily measured through credit utilization, the percentage of available credit being used. A $3,000 balance on a $10,000 limit is 30% utilization. Lower utilization correlates with higher scores, with significant drops typically occurring above 30% and steeper drops above 50%. This factor also considers the number of accounts with balances and how much of installment loans has been paid down.
Length of credit history (15%): The age of the oldest account, the age of the newest account, and the average age across all accounts. Older credit history signals more data for the model to evaluate. Closing old accounts or opening many new ones reduces average age. This factor is why people maintain old credit cards even when unused, the account age continues benefiting their score.
Credit mix (10%): The variety of account types, including credit cards (revolving), mortgages, auto loans, student loans, and personal loans (all installment). Having experience with different credit types indicates broader credit management capability. This doesn’t mean opening unnecessary accounts, but it does mean someone with only credit cards might score slightly lower than someone with cards plus an installment loan.
New credit (10%): Recent applications and new accounts opened. Each application for credit triggers a “hard inquiry” that temporarily reduces scores by a few points. Multiple inquiries in a short period for the same loan type (mortgage, auto) are typically grouped and treated as single inquiry, since rate shopping is expected. New accounts reduce average account age and initially have no payment history, both minor negatives.
How utilization actually affects scores
Credit utilization deserves deeper explanation because it’s the most actionable factor for short-term score changes.
Utilization is calculated at both the individual card level and across all revolving accounts. If someone has three cards with limits of $10,000, $5,000, and $5,000 ($20,000 total), a $6,000 balance on the first card produces 60% individual utilization on that card and 30% overall utilization.
Both individual and aggregate utilization matter, though overall utilization carries more weight. Maxing out one card while others sit at zero looks worse in scoring models than spreading the same total balance across cards.
Utilization has no memory. The score only reflects the most recently reported balances. This creates short-term optimization opportunities: paying down balances before statement closing dates reduces the utilization that gets reported to bureaus. A card used heavily mid-cycle but paid before the statement closes reports low utilization.
The utilization thresholds that appear in scoring guidance, keep below 30%, ideally below 10%, are approximations based on observed score patterns. The actual algorithm is continuous rather than stepped, meaning utilization affects scores across the entire range, not just at specific thresholds.
What doesn’t affect scores
Several things people believe affect credit scores actually don’t:
Income: Credit reports don’t contain income information. A person earning $500,000 and a person earning $50,000 with identical credit behaviors would have identical scores. Lenders consider income separately when evaluating applications.
Employment: Job history doesn’t appear on credit reports beyond employer names used for identification purposes. Unemployment doesn’t directly affect scores, though the inability to pay bills during unemployment certainly does.
Checking and savings accounts: Bank accounts aren’t part of credit reports. Account balances, overdrafts, and banking history don’t factor into credit scores.
Rent payments: Traditionally not reported, though some services now enable rent reporting. If not reported, rent payment history doesn’t affect scores.
Debit card usage: Debit transactions don’t appear on credit reports. Using a debit card builds no credit history.
Soft inquiries: Checking your own credit, employer background checks, and pre-approved offers trigger “soft inquiries” that don’t affect scores. Only “hard inquiries” from actual credit applications impact scores.
How scores respond to changes
Credit scores update when new information appears on credit reports. Most creditors report monthly, typically around statement closing dates. After new data reaches the bureaus, scores recalculate within a few days.
This means score changes have a lag of roughly 30-45 days from action to reflection. A balance paid on March 1 might not report until the March 15 statement close, might not reach bureaus until March 20, and might not show in scores until late March.
Different types of changes affect scores at different speeds:
Fast response: Credit utilization changes reflect within one reporting cycle. Paying down a high balance or opening a new card with available credit can shift scores within 30-45 days.
Medium response: Adding new positive accounts takes a few months to show meaningful impact. The account needs time to build payment history.
Slow improvement: Recovering from serious negatives (late payments, collections, bankruptcies) takes years. The impact of a late payment diminishes gradually but remains on reports for seven years. Bankruptcies remain for seven to ten years depending on type.
Score ranges and what they mean
FICO scores range from 300 to 850. VantageScores also use 300-850 for current versions. The distribution isn’t uniform; most scores cluster between 650 and 750, with relatively few at the extremes.
Excellent (750+): Qualifies for the best interest rates and terms on most products. Lenders see minimal risk. Above 800 provides no additional benefit, as the best rates are already available.
Good (700-749): Qualifies for competitive rates, though not always the absolute best. Most prime lending products are accessible.
Fair (650-699): Qualifies for many products but with higher rates. Some premium products may be unavailable. Subprime rates begin appearing.
Poor (550-649): Limited options, high rates, and often requires secured products or cosigners. Many mainstream lenders decline applications in this range.
Very Poor (below 550): Most traditional credit products are unavailable. Options are largely limited to secured cards, credit-builder loans, and predatory products.
These categories are approximations. Different lenders use different cutoffs, and other factors (income, employment, existing relationship) influence decisions alongside scores.
Where scores come from
Three credit bureaus collect and maintain credit data: Equifax, Experian, and TransUnion. Each bureau independently gathers information from creditors, public records, and other sources. They then sell this data to lenders and sell scores calculated from it.
Consumers can access their credit reports for free at AnnualCreditReport.com, the only federally authorized source for free annual reports. Each bureau provides one free report per year, and recent rules allow free weekly access.
Credit reports contain the raw data. Credit scores are calculated from that data. Getting your report doesn’t automatically give you your score, they’re separate products. Many banks and credit cards now provide free FICO scores to customers, and services like Credit Karma provide free VantageScores.
Disputing errors
Credit reports can contain errors: accounts that aren’t yours, incorrect payment statuses, outdated information that should have aged off. Federal law (Fair Credit Reporting Act) gives consumers the right to dispute errors and requires bureaus to investigate within 30 days.
Disputes can be filed online through each bureau’s website or by mail. Providing documentation supporting the dispute, statements showing on-time payments, proof of identity theft, increases the chance of successful removal.
Errors that significantly affect scores are worth disputing. Minor errors (wrong employer name, slightly off credit limits) don’t affect scores and may not be worth the effort. The goal is accuracy in the data that actually feeds scoring models.
The relationship between scores and rates
Credit scores directly affect borrowing costs. On a $300,000 30-year mortgage, the difference between a 720 score qualifying for 6.5% and a 680 score paying 7.0% is roughly $100/month, $36,000 over the loan’s life.
Auto loans, credit cards, insurance premiums, and even rental applications use credit scores to set pricing or make approval decisions. The cumulative cost of lower scores over a lifetime can reach six figures in additional interest and fees.
This doesn’t mean obsessing over every point. The difference between 750 and 780 is negligible in most lending decisions. But the difference between 680 and 720 is meaningful, and the difference between 620 and 680 can determine approval versus denial.
What scores don’t capture
Credit scores measure past credit management as a predictor of future default risk. They don’t measure financial health more broadly. Someone with a high score might have no savings, unsustainable spending, or precarious income. Someone with a low score might have substantial assets but past credit problems.
Scores are one tool lenders use, not the only tool. Income verification, employment history, down payments, and existing assets all factor into lending decisions. A strong score with no income won’t get approved. A weak score with substantial assets might find manual underwriting paths available.
Understanding what scores are, how they work, and what they’re used for puts the number in proper context. It matters, but it’s not the whole picture of financial life.