Borrower's Guide
General information, not legal or financial advice. Sources are named on every page.

Credit Score Ranges: What 'Good' Actually Means

Credit score ranges run from 300 to 850, but 'good' is set by each lender's cut-offs. Learn how FICO and VantageScore ranges work and why scores differ.

A credit score is a number produced by a scoring model from the contents of a credit file. The ranges most consumers encounter run from 300 to 850. That range is a convention of the two dominant model families, FICO and VantageScore, not a feature of federal law. No statute defines a “good” score. Lenders set their own cut-offs, and those cut-offs vary by product, by lender, and over time.

The Consumer Financial Protection Bureau (CFPB) publishes an explainer on credit scores that describes the general mechanics: payment history, amounts owed, length of credit history, new credit, and credit mix. The CFPB does not publish a national threshold for approval. The FICO and VantageScore organizations publish score ranges and, in FICO’s case, descriptive labels for bands within the range. Those labels are the model vendor’s own; they are not regulatory categories.

The published ranges

Base FICO Scores run from 300 to 850. VantageScore 3.0 and 4.0 also run from 300 to 850. Some industry-specific FICO versions use a different range: FICO Auto Scores and FICO Bankcard Scores, for example, are scaled differently from the base score, so a number from one of those models is not directly comparable to a base FICO Score. The same applies to any model whose documentation states a different scale.

FICO publishes descriptive bands within its 300–850 range. Those bands carry labels such as “poor,” “fair,” “good,” “very good,” and “exceptional.” They are FICO’s own labels for describing a score, not a legal standard and not a promise about any lender’s decision. VantageScore publishes its own descriptive language for its range. Neither vendor’s labels bind a lender.

Model or sourceRangeWhat the range meansWhen it does not apply
Base FICO Score300–850The scale used for most general-purpose FICO scoresNot comparable to industry-specific FICO versions with a different scale
FICO Auto / Bankcard ScoresDifferent scale (see model documentation)Industry-specific scores used in those lending decisionsNot interchangeable with a base FICO Score
VantageScore 3.0 / 4.0300–850A separate model family used by some lenders and consumer sitesNot the same model as FICO; a VantageScore is not a FICO Score
Lender cut-offSet by the lenderThe threshold that lender applies to that productNot a national standard; varies by lender and product

Why there is no single score

A credit score is the output of a model applied to a credit file. Three nationwide consumer reporting agencies hold files: Equifax, Experian, and TransUnion. The files are not identical, because creditors report to different agencies and on different schedules. A single model applied to three different files can produce three different scores. Different models applied to the same file can also produce different scores, because the models weight the data differently.

FICO and VantageScore are separate model families, each with multiple versions. A lender chooses which model and version to use for a given product. A mortgage lender, an auto lender, and a credit card issuer may each use a different FICO version. A consumer who sees a score in a banking app or on a consumer site is seeing the output of whichever model that service uses, applied to whichever file it has access to. That score may differ from the score a lender pulls, because the lender may use a different model, a different version, or a different agency’s file.

The CFPB explainer notes that scores can differ across sources and that consumers may see more than one score. The existence of multiple scores is a structural feature of the system, not an error.

No federal statute sets a minimum credit score for approval of a loan or a credit card. Each lender sets its own underwriting criteria, which may include a score cut-off, and those criteria can change. A score that one lender treats as acceptable may be below another lender’s cut-off for the same product. A score that is acceptable for one product may be below the cut-off for another.

The tier labels — “good,” “fair,” “poor” — are conventions used by model vendors and consumer sites to describe where a number falls within a range. They are useful for orientation and nothing more. A consumer whose score falls in a band labeled “good” by one vendor may still be declined by a lender whose cut-off is higher, and a consumer whose score falls in a lower band may still be approved by a lender with a lower cut-off or with other compensating factors in the file.

Myths that drive most searches

One universal score. There is no single credit score. There are multiple models, multiple versions, and three agency files. A score is a model output tied to a file at a point in time. Any statement that a person “has a 720” is shorthand for a score from a particular model on a particular file on a particular date.

Checking your own score lowers it. Checking a consumer’s own score is a soft inquiry. Soft inquiries are not used in scoring models the way hard inquiries are. A hard inquiry results when a lender pulls a file in connection with an application for credit, and hard inquiries can affect a score. The distinction is between a consumer-initiated check and a lender-initiated application pull. The CFPB explainer describes the difference between soft and hard inquiries.

A score is a permanent attribute. A score changes as the file changes. Balances, new accounts, inquiries, and the passage of time all move the inputs. The score is a snapshot, not a fixed trait.

The mistake people most often make

The most common mistake is treating a score from one source as the definitive number that every lender will see. It happens because the number is presented without its context: an app shows a score, the number looks authoritative, and the model and file behind it are not visible. A consumer may then compare that number to a published band, conclude that the number is “good,” and be surprised when a lender declines an application. The decline may reflect a different model, a different version, a different agency’s file, or a lender cut-off above the band the consumer was reading. The error is not in checking the score; it is in assuming that one score is the score.

A related mistake is assuming that the descriptive bands published by a model vendor are the same as lender cut-offs. They are not. The vendor’s bands describe the range; the lender’s cut-off determines the decision.

What moves a score

FICO publishes the weightings of the factors in its base score. Payment history carries the largest share, followed by amounts owed (which includes credit utilization), length of credit history, new credit, and credit mix. The exact percentages are published by FICO and are structural definitions of the model, not prices. VantageScore publishes its own factor descriptions. The CFPB explainer describes the same general categories.

Credit utilization is the balance on revolving accounts relative to the credit limits. It is a ratio, and it is one of the inputs in the amounts-owed category. Utilization thresholds are sometimes described in consumer guidance, but the model’s treatment of utilization is a model parameter, not a statutory rule.

Where the law comes in

The Fair Credit Reporting Act (FCRA), 15 U.S.C. § 1681 et seq., governs consumer reporting agencies and the users of consumer reports. It gives consumers the right to dispute inaccurate information in their files and imposes obligations on agencies and furnishers. It does not set a minimum score for credit approval. The CFPB supervises consumer reporting agencies and publishes consumer-facing explanations of scoring and reporting. State laws may add requirements, and the details of dispute procedures and timelines can vary by state; a consumer facing a specific dispute should check the procedure that applies in that state.

A lender’s adverse action notice, required in certain circumstances under the Equal Credit Opportunity Act (ECOA) and its implementing regulation, Regulation B, 12 C.F.R. Part 1002, may state the reasons for a denial, which can include the score or the factors that contributed to it. The notice describes the lender’s decision; it does not establish a national threshold.

What to look up

A consumer who wants to understand a specific score should identify the model and version behind it, the agency whose file was used, and the date. The model vendor’s documentation states the range and the factor weightings. The CFPB explainer describes the general mechanics and the difference between soft and hard inquiries. For a denial, the adverse action notice states the reasons the lender gave. For a dispute about file contents, the FCRA and the relevant state procedure govern. Where a denial or a dispute involves a large sum or a lawsuit, that is the point to get actual legal help.

Common questions

What is a good credit score?

There is no national definition. FICO and VantageScore publish descriptive bands within their 300–850 ranges, but each lender sets its own cut-off for approval. A score that one lender treats as acceptable may be below another lender's cut-off for the same product.

What is the FICO score range?

Base FICO Scores run from 300 to 850. Some industry-specific FICO versions, such as FICO Auto Scores and FICO Bankcard Scores, use a different scale and are not directly comparable to a base FICO Score.

What is the VantageScore range?

VantageScore 3.0 and 4.0 run from 300 to 850. VantageScore is a separate model family from FICO, so a VantageScore is not the same as a FICO Score even when both fall on the same numerical range.

Does checking my own credit score lower it?

Checking your own score is a soft inquiry, and soft inquiries are not treated the same as hard inquiries in scoring models. A hard inquiry results when a lender pulls your file in connection with an application for credit.

Why is my credit score different in different apps?

Different apps may use different scoring models, different versions of a model, or files from different credit reporting agencies. A score is a model output tied to a specific file at a specific time, so differences across sources are expected.

Read next