James Monroe had a 790 credit score when he applied for an apartment in Denver. The landlord's tenant screening service returned results he had never seen before: his score, a summary of his credit history, and, at the bottom, a line he had not anticipated. The service flagged that he had no rental history at the Denver address tier. The landlord declined his application. James had excellent credit by any conventional measure, but the screening model used a specialty score that incorporated rental history separately from FICO, and his clean slate read as risk.

James's experience illustrates something that most people who think about credit scores do not fully understand: a credit score is not a single number computed by a neutral authority from objective facts. It is a model output, one of several competing models, each using proprietary formulas applied to data supplied by one of three private bureaus, each of which collects data independently and makes its own decisions about what to include. The system is consequential, affecting mortgage rates, rental approvals, insurance premiums, and in some contexts employment eligibility, and its mechanics are less transparent than its influence suggests they should be.

Understanding how the system works, what moves scores up and down, and where the documented failure points are, is not merely useful for individuals navigating credit decisions. It is also a necessary map for anyone trying to build financial stability in a society where a three-digit number has significant leverage over major life outcomes.

"Your credit score is not a measure of your financial health. It is a measure of your value as a debt customer. These are related but not identical things." -- Ann Carrns, personal finance reporter, The New York Times, 2019


Key Definitions

FICO score: The credit score produced by Fair Isaac Corporation (now branded simply as FICO), first introduced in 1989. Used in approximately 90 percent of US consumer lending decisions, according to FICO's own reporting. Scores range from 300 to 850. FICO has multiple versions (FICO 8, FICO 9, FICO 10, industry-specific versions for mortgage and auto lending) which can produce different scores from the same underlying data.

VantageScore: A competing credit scoring model developed jointly by the three major credit bureaus (Equifax, Experian, TransUnion) and launched in 2006. Also scored on a 300 to 850 scale. More commonly used for soft-pull and pre-qualification purposes and by some non-lending applications (landlords, insurance, employment) than for mortgage origination, where FICO dominates.

Credit bureau: One of the three companies (Equifax, Experian, TransUnion) that collect credit information from lenders, compile credit reports, and supply report data to scoring models. Each bureau collects data independently; not all lenders report to all three bureaus; the same person's report at the three bureaus may differ.

Hard inquiry: A credit inquiry generated when a consumer applies for credit. Lenders perform hard pulls to evaluate creditworthiness for loan or credit card applications. Hard inquiries appear on credit reports and can reduce scores by a small amount for up to 24 months.

Soft inquiry: A credit inquiry that does not affect credit scores. Includes the consumer checking their own credit, pre-approval offers from lenders, and most background checks for employment or rental screening.

Credit utilization: The ratio of current revolving credit balances to total available revolving credit limits, expressed as a percentage. A balance of $2,000 on cards with a combined limit of $10,000 equals 20 percent utilization.

Thin file: A credit file with insufficient history for a scoring model to generate a reliable score, typically fewer than three open accounts or less than six months of credit history.


FICO Score Factors and Their Weights

Factor Weight What It Measures Key Action
Payment history 35% On-time vs. late payments across all accounts Never miss a payment; set up autopay for minimums
Credit utilization 30% Revolving balance as % of total credit limit Keep utilization below 30%; ideally under 10%
Length of credit history 15% Age of oldest account, newest account, and average age Keep old accounts open; avoid unnecessary new accounts
Credit mix 10% Variety of account types (credit cards, installment loans, mortgage) Diversify types of credit over time
New credit inquiries 10% Hard inquiries from recent credit applications Limit applications; rate-shop within a 45-day window

The FICO Model: Five Factors and Their Weights

Fair Isaac Corporation has publicly disclosed the approximate relative importance of the five factors that contribute to FICO scores. The weights are not exact formulas (those are proprietary) but describe the broad contribution of each category:

Payment History: 35 Percent

The single largest contributor to a FICO score is whether you have paid your bills on time. Every account reported to the bureaus contributes to this factor: credit cards, installment loans, mortgages, home equity lines, and certain utilities and rental payments if reported.

The impact of a missed payment is immediate and lasting. A payment that is 30 days late is reported as a derogatory mark. A single 30-day late payment on an otherwise clean record can reduce a score by 50 to 100 points, depending on the starting score and overall file characteristics. Scores above 750 tend to show larger drops from a single late payment because there is more room to fall and the record of consistent payment makes the deviation more statistically notable.

The timeline for recovery: a single late payment's impact fades substantially by two years and the item drops off entirely after seven years from the original delinquency date. More severe derogatory marks -- charge-offs, collections, foreclosure, bankruptcy -- follow the same seven-year timeline (ten years for Chapter 7 bankruptcy).

The practical implication is unambiguous: no financial optimization of any other credit factor is worth the cost of a single missed payment. Set automatic minimum payments on every account to prevent inadvertent late payments.

Credit Utilization: 30 Percent

Credit utilization is the second most important factor and one of the most actionable, because it has no memory: paying down a balance this month is immediately reflected in next month's score, regardless of what utilization was in previous months.

The scoring models evaluate both total utilization across all revolving accounts combined and individual account utilization. Carrying $9,500 on a single card with a $10,000 limit damages the score even if no other card has any balance.

Published guidelines from FICO and credit counseling organizations recommend keeping utilization below 30 percent for good score maintenance. Research and analyses by credit scoring practitioners suggest below 10 percent is optimal for the highest scores. Data from borrowers in the 800+ score range shows average utilization typically below 7 percent, though causation and correlation are intertwined here -- people with excellent scores tend to have both high credit limits (reducing utilization mathematically) and low balances.

The statement-close timing insight: Most credit card issuers report balances to credit bureaus at statement close, not at payment due date. If you pay your balance in full every month but the statement closes when your balance is high, the reported utilization is high regardless of the fact that you paid it. Paying the balance before the statement closes, rather than just before the payment due date, reduces reported utilization. This matters most when you plan to apply for new credit in the near future.

Length of Credit History: 15 Percent

This factor rewards having older accounts and a longer average account age. Closing old credit card accounts, even ones you are no longer using, reduces your average account age and can hurt scores. The prevailing advice is to keep old accounts open even if unused, provided they have no annual fee. If the account has an annual fee, the trade-off requires calculation.

The length of history factor is one reason building credit takes time by design: there is no shortcut to having accounts that have existed for ten years except waiting ten years. Becoming an authorized user on an older account can add the account's history to your file, though bureau policies vary and some scoring models weight authorized user accounts differently than primary accounts.

Credit Mix: 10 Percent

Credit scoring models reward having experience with different types of credit: revolving accounts (credit cards, home equity lines of credit) and installment accounts (mortgages, auto loans, student loans, personal loans). The research rationale is that managing different types of credit successfully provides additional predictive signal about repayment behavior.

This factor should not be chased actively. Taking out an installment loan you do not need to improve your credit mix is counterproductive. The factor rewards naturally diverse credit use over time, not strategic borrowing for score optimization.

New Credit: 10 Percent

New credit accounts for 10 percent of the FICO score through two mechanisms: hard inquiries (each application for credit generates one) and the number of recently opened accounts. Both indicate increased credit-seeking activity, which scoring models correlate with higher default risk.

The hard inquiry impact is modest: typically 5 points or fewer per inquiry, fading within 12 months, disappearing entirely at 24 months. The rate-shopping exception is important: FICO and VantageScore both recognize that consumers legitimately compare rates for mortgages, auto loans, and student loans. Multiple inquiries for the same loan type within a 14 to 45 day window (the window varies by FICO version) are counted as a single inquiry in the scoring calculation.

VantageScore: FICO's Competitor

VantageScore was created jointly by the three major credit bureaus in 2006 and has steadily gained market share, particularly in non-mortgage lending and in consumer-facing score products. The major credit monitoring services (Credit Karma, CreditWise from Capital One) typically display VantageScore 3.0, which is why scores visible to consumers in these apps often differ from FICO scores pulled by lenders.

VantageScore 4.0, the current version, uses many of the same underlying factors as FICO but weights them differently and uses more nuanced algorithms for thin files. VantageScore can score some consumers with fewer than six months of credit history that FICO cannot score, making it useful for consumers new to credit.

The practical implication: the score you see on Credit Karma is likely a VantageScore, which is informative about your credit standing but may differ from the FICO score a mortgage lender will pull. If you are preparing for a major credit application, checking a FICO score (available through myFICO.com for a fee, and free through some credit card benefits programs) gives a more accurate preview.

Credit Report Errors: The FTC Research

The Federal Trade Commission conducted a comprehensive study, published in 2013, examining the accuracy of consumer credit reports. The findings were sobering: approximately 1 in 5 (20 percent) of consumers had at least one error in their credit reports that was significant enough to potentially affect their credit score. Approximately 1 in 20 consumers had an error serious enough to cause them to receive a less favorable credit term from a lender.

Common error types include:

  • Accounts belonging to someone else with a similar name or Social Security number
  • Accounts the consumer did not open (potential identity theft or data error)
  • Payments reported as late that were actually made on time
  • Accounts showing open and with balances that the consumer had paid off and closed
  • Duplicate accounts listed multiple times
  • Incorrect personal information (wrong address, employment history)

Disputing errors: Under the Fair Credit Reporting Act (FCRA), consumers have the right to dispute inaccurate information. Disputes must be submitted to the bureau reporting the error (not to the creditor), and the bureau has 30 days to investigate. The process is free. If the investigation confirms the error, the information must be corrected or deleted. Disputes can be filed at each bureau's website or by mail.

The free credit report service at AnnualCreditReport.com (the federally mandated site) now provides weekly access to all three bureau reports, upgraded from annual access during the COVID-19 period. Reviewing all three reports annually for errors is basic financial hygiene.

Building Credit from Scratch: The Thin File Problem

An estimated 45 million Americans have no credit score, either because they have no credit history (thin file) or because their file has become too inactive to score. Recent immigrants, young adults, people who have only used cash and debit, and people who have avoided formal banking are disproportionately represented.

The thin file problem creates a genuine paradox: you need credit history to get credit, but you cannot get credit without history. The standard solutions:

Secured credit cards: A credit card backed by a cash deposit equal to the credit limit. If you deposit $500, your credit limit is $500. The deposit protects the issuer, allowing approval for people with no credit history. Capital One Secured, Discover it Secured, and the Chime Credit Builder Visa (which works differently, requiring no minimum deposit) are commonly recommended. Using the card for small purchases and paying the balance in full monthly for 6 to 12 months typically generates a scoreable file and begins building payment history.

Credit-builder loans: Offered by credit unions, community banks, and services like Self Financial (formerly Self Lender). The borrower makes monthly payments into a savings account held by the lender; the lender reports the payments to credit bureaus. At the end of the loan term, the borrower receives the accumulated savings (minus fees and interest). The credit-building benefit comes entirely from the reported payment history.

Becoming an authorized user: If a family member or trusted person has a long-standing, well-managed credit card account, being added as an authorized user can add that account's history to your credit file. You do not need to use the card. The benefits vary by bureau and scoring model version.

UltraFICO and Experian Boost: Alternative data programs that allow consumers to opt in to having non-traditional payment history included in score calculations. Experian Boost allows adding on-time utility and telecom payments; UltraFICO incorporates bank account management patterns. These programs help thin-file consumers most and can raise scores meaningfully for people with limited traditional credit history.

Credit Freezes and Locks

A credit freeze (also called a security freeze) prevents any new credit inquiry from accessing your credit report. It is the strongest protection against identity theft-based new account fraud. Freezing your report does not affect your credit score and does not prevent you from using existing accounts.

Under federal law since 2018, credit freezes are free to place and lift at all three bureaus. Unfreezing (lifting) a freeze is required before applying for new credit and typically takes minutes to hours online. The freeze can also be temporarily lifted for a specific date range when you anticipate applying for credit.

A credit lock is a similar product offered commercially by the credit bureaus themselves. Locks are faster to toggle on and off than freezes and may include monitoring features, but they are contractual agreements rather than legal rights. The legal protection of a statutory credit freeze is generally considered stronger.

For people who are not actively applying for new credit, placing a freeze at all three bureaus is a straightforward protective measure against the most common form of identity theft.

How Credit Scores Affect Your Financial Life

Mortgage Rates

The difference between a 620 FICO score and a 760 FICO score on a 30-year fixed mortgage is substantial. Myfico.com's loan savings calculator has historically shown interest rate differences of 1.0 to 1.5 percentage points between these score ranges. On a $300,000 mortgage, 1.0 percentage point difference in rate equals approximately $60 more per month, or approximately $21,600 over a 30-year loan term.

Insurance Premiums

Most US states allow insurers to use credit-based insurance scores (distinct from lending scores but based on the same underlying bureau data) to price auto and homeowner insurance. Drivers with poor credit scores pay, on average, 52 to 91 percent more for auto insurance than drivers with excellent credit, according to research by the Consumer Federation of America. Seven states (California, Hawaii, Maryland, Massachusetts, Michigan, Oregon, Washington) prohibit the use of credit in auto insurance pricing.

Rental Screening

Credit checks have become nearly universal in rental applications in most US cities. Landlords and property management companies use a range of tools, from manual report review to automated scoring services. The specific score thresholds vary by landlord, but most large property management companies have defined minimum FICO thresholds (commonly 620 to 650) and will decline applicants below that threshold regardless of income or other factors.

Employment

As noted in the FAQ section, employer credit checks are legally permitted in most US states with written consent. The research on whether credit history predicts job performance is mixed at best: a 2014 analysis by Jeremy Bernerth of Louisiana State University found minimal predictive validity for employment credit checks in most job categories. The practice is more entrenched in practice than its evidence base justifies.

Score Improvement: A Practical Sequence

For someone starting from a damaged credit score (below 620) with some derogatory history:

  1. Pull all three credit reports from AnnualCreditReport.com and verify every entry for accuracy. Dispute all factual errors.
  2. Set automatic minimum payments on every open account to prevent any future late payments while the remediation process proceeds.
  3. Address any accounts in collections: For older collections, consult with a credit counselor before paying -- paying some old collections can restart the seven-year clock on newer scoring models, a counterintuitive result. For recent collections (within the past two years), paying or settling removes the ongoing damage from the open derogatory mark.
  4. Pay down revolving balances to below 30 percent of the credit limit on each card. Then below 10 percent on each card if possible.
  5. Do not close old accounts during the remediation period.
  6. Do not apply for new credit unless necessary during active score repair, as hard inquiries add marginal negative impact.
  7. Monitor progress monthly using a free service (Experian free tier, Credit Karma, Capital One CreditWise) to track score trajectory.

The realistic timeline for moving from 580 to 680 with consistent positive behavior and no new negative items is 12 to 24 months. Moving from 680 to 750 typically takes another 12 to 24 months of continued clean history. Scores above 750 are primarily a function of long, consistent positive history with low utilization, and the timeline reflects that.


References

  1. Fair Isaac Corporation. (2024). What's in my FICO Scores? MyFICO.com.
  2. Federal Trade Commission. (2013). Report to Congress Under Section 319 of the Fair and Accurate Credit Transactions Act of 2003. FTC.gov.
  3. Consumer Financial Protection Bureau. (2023). Consumer Credit Reports: A Study of Medical and Non-Medical Collections. CFPB.gov.
  4. VantageScore Solutions. (2024). How VantageScore 4.0 Works. VantageScore.com.
  5. Consumer Federation of America. (2020). Credit Scores: What They Mean for Consumers. ConsumerFed.org.
  6. Bernerth, J. B., Taylor, S. G., Walker, H. J., & Whitman, D. S. (2012). An empirical investigation of dispositional antecedents and performance-related outcomes of credit scores. Journal of Applied Psychology, 97(2), 469-478.
  7. Chi, G., Qian, Z., & Qian, H. (2020). Disparate impacts of credit scoring on minority communities. Housing Policy Debate, 30(5), 785-803.
  8. Federal Trade Commission. (2018). Free Credit Freezes Now Available. FTC Consumer Information.
  9. Experian. (2024). Experian Boost: How It Works. Experian.com.
  10. CFPB. (2022). Consumer Financial Protection Bureau Supervisory Highlights: Credit Reporting. CFPB.gov.
  11. Avery, R. B., Brevoort, K. P., & Canner, G. B. (2009). Credit scoring and its effects on the availability and affordability of credit. Journal of Consumer Affairs, 43(3), 516-537.
  12. Sullivan, T., Warren, E., & Westbrook, J. L. (2000). The Fragile Middle Class: Americans in Debt. Yale University Press.

Related reading: how to save money effectively, how to start investing, why people make bad financial decisions, what is behavioral finance

Frequently Asked Questions

What factors affect your credit score the most?

The FICO score, used in approximately 90 percent of US lending decisions, weights five factors in specific proportions. Payment history is the most important at 35 percent of the score: whether you pay every bill by the due date, every month, without exception. A single missed payment of 30 days or more can drop a score by 50 to 100 points. Credit utilization is the second most important factor at 30 percent: the ratio of your current revolving credit balances to your total credit limits. Using \(3,000 of a \)10,000 total credit limit means 30 percent utilization. Credit scoring models reward lower utilization rates, with below 30 percent generally positive and below 10 percent optimal. Length of credit history accounts for 15 percent, rewarding longer average account ages. Credit mix accounts for 10 percent, rewarding a combination of revolving accounts (credit cards) and installment accounts (auto loans, mortgages, student loans). New credit accounts for 10 percent, including the number of hard inquiries and recently opened accounts. The FICO model is proprietary and the exact calculation formulas are trade secrets, but these factor weights represent FICO's own public disclosure of their approximate importance.

How do you build credit from scratch?

People with no credit history, sometimes called having a 'thin file,' face the classic paradox: lenders require credit history to extend credit, but you need credit to build history. The standard solutions are: a secured credit card (requires a cash deposit equal to the credit limit, which protects the lender; Capital One Secured, Discover it Secured, and Chime Credit Builder are common options); a credit-builder loan (offered by credit unions and community banks, including Self Financial and many credit unions, where you make payments into a savings account that you receive at the end, while the payment history is reported to the bureaus); becoming an authorized user on a family member's established account (the account's history can appear on your report, though policies vary by bureau and card issuer); and student credit cards for eligible college students. The timeline for a thin file to become scoreable is typically 3 to 6 months of account activity. Moving from a thin file to a good credit score (700 or above) typically takes 12 to 24 months of consistent on-time payments with low utilization. The most important single action is making every single payment on time: payment history is 35 percent of the score.

What credit utilization percentage is best?

Credit scoring models reward lower utilization consistently. The general guideline widely cited by credit professionals is to keep total utilization below 30 percent. For the best scores, below 10 percent is optimal. This means on a \(10,000 total credit limit across all cards, carrying a balance of under \)1,000 at the time scores are calculated (which typically occurs when the statement closes, not the payment due date). Two points that are not widely understood: first, utilization has no memory in FICO models. Paying down a high balance fully one month produces the same utilization benefit as having maintained a low balance all along. Second, utilization applies both to total credit limit across all cards combined and to individual card utilization. A card at 90 percent utilization hurts the score even if total across all cards is low. If you plan to apply for a major loan (mortgage, auto), paying down card balances to below 10 percent in the month before the application can produce a meaningful score improvement. The utilization calculation uses the balance reported at statement close, so paying before the statement closes (not just before the due date) is the method to control reported utilization.

How long does it take to improve a bad credit score?

It depends on what caused the damage and when it occurred, because different negative items have different timelines. A single late payment (30 days) hurts for approximately two years before its impact fades significantly, and drops off the report entirely after seven years. Multiple missed payments, a charge-off, or a collection account: seven years from the original delinquency date. Bankruptcy: Chapter 7 stays for ten years; Chapter 13 stays for seven years. A foreclosure: seven years. Hard inquiries: two years. The good news is that recent behavior matters more than old behavior in credit scoring models: a collection from five years ago with perfect payment history since then has limited impact on current scores. Active positive behavior, consistent on-time payments and low utilization, produces score improvement regardless of what is in the historical record. Most people who start from 550 to 600 with consistent positive behavior reach 680 to 700 within 12 to 24 months without any of the old negative items being removed. Disputes of legitimate negative items rarely succeed; the proper dispute mechanism is for information that is factually inaccurate.

Does checking your own credit score hurt it?

No. Checking your own credit score is a soft inquiry, which has no impact on credit scores. Soft inquiries include checking your own credit, pre-approval checks by lenders, and employment background checks. Hard inquiries, which can affect scores, occur only when you formally apply for credit: a credit card application, a mortgage pre-approval, an auto loan application, a personal loan. A single hard inquiry typically reduces a FICO score by 5 points or less, and the impact fades within 12 months and disappears entirely after 24 months. The exception to the 'one application, one inquiry' rule is rate shopping: FICO and VantageScore both treat multiple mortgage, auto, or student loan inquiries within a short window (14 to 45 days depending on the model version) as a single inquiry, recognizing that consumers legitimately shop for the best rate. You can check your own credit reports for free at AnnualCreditReport.com (the official site mandated by federal law), now providing weekly access to all three bureau reports.

What are the biggest mistakes that hurt credit scores?

The highest-impact mistakes are: missing payments by 30 days or more, which triggers a derogatory mark that persists for seven years; maxing out credit cards or carrying balances near the credit limit, since high utilization is the second most important factor and is immediately visible to scoring models; closing old credit card accounts, which can hurt in two ways -- it reduces total available credit (increasing utilization on remaining cards) and may lower average account age; applying for multiple new credit accounts in a short period, which generates multiple hard inquiries and also lowers average account age; cosigning a loan for someone who then misses payments (you are equally responsible for the debt and all derogatory information appears on your report too); and ignoring a credit report error, since FTC research found that one in five consumers has a material error on at least one bureau report. The error rate finding means everyone should periodically verify their reports. Errors can include accounts that do not belong to you (due to fraud or someone with a similar name), payments reported as missed that were actually made, and balances reported higher than actual.

Can landlords and employers really check your credit?

Yes, both can with your permission, though what they access and what they use it for differs from lender use. Landlords commonly pull credit reports as part of tenant screening; many use third-party services (TransUnion SmartMove, Experian RentBureau) that provide a specialized report for rental purposes. Landlords look primarily for prior evictions, patterns of late payment, outstanding collections, and bankruptcy. They typically see a report rather than a specific score, though some landlord-facing products do provide a score. Employers in approximately 47 US states may check credit reports for employment purposes, though only 13 states have enacted restrictions on the practice. Employers must obtain written consent before pulling a credit report and must follow specific notification procedures if the report influences an employment decision, under the Fair Credit Reporting Act. Employers see a modified version of the credit report that omits the credit score, birth year, and account numbers. Research by Chi et al. and others has raised disparate impact concerns: credit history correlates with race and socioeconomic background in ways that may have discriminatory effects in employment screening. Seven US states (California, Colorado, Connecticut, Hawaii, Illinois, Maryland, Oregon, Washington) restrict employer use of credit reports more broadly.