Hard Inquiry vs Soft: What Hits Your Score

There is a small, specific dread that comes with the moment before you submit a credit application. You have compared the offers, you have read the fine print, and your finger hovers over the button because somewhere you absorbed the idea that clicking it will damage your credit. That dread is mostly unearned. The thing you are afraid of is a hard inquiry, and for the overwhelming majority of borrowers it is the smallest, fastest-fading mark that can appear on a credit report.

This playbook takes the inquiry apart. It covers what a hard inquiry is and who is allowed to make one, how the hard and soft versions differ with concrete examples of each, how far a pull actually moves a score and for how long, why several mortgage or auto pulls are commonly treated as one event while card applications are not, how to find the inquiries sitting on your file right now, and what you can and cannot do about one you dislike. Where it borders on our notes about how credit utilization works and why a credit score goes down, it points there rather than repeating them.

What a hard inquiry actually is

A hard inquiry, sometimes called a hard pull, is the footprint left when a company asks a credit bureau for your file in connection with a decision about extending you credit. It is not a judgment, a penalty, or a flag. It is a log entry: this company, on this date, looked at this report. The bureau records it, keeps it attached to your file, and shows it to anyone who later pulls the same report.

The reason it carries any scoring weight at all is that it is evidence of intent. A lender considering your application wants to know whether you are also asking three other lenders for money this week, because someone reaching for credit in several directions at once is, on average, in a different situation than someone who is not. The inquiry is the only trace of an application that exists before an account opens, so scoring models use it as a small early signal.

Two things follow from that framing, and both are reassuring. First, the signal is weak by design, because plenty of people apply for credit for perfectly boring reasons. Second, it is short-lived, because intent from eighteen months ago says almost nothing about you today. A hard inquiry is the closest thing a credit report has to a footnote.

Hard inquiry vs soft inquiry: the real difference

The clean way to hold the distinction is to ask who initiated the look and why. A hard inquiry happens when you ask a company for credit and it pulls your file to decide. A soft inquiry happens for every other reason: you checking yourself, a company checking you for marketing, an existing lender reviewing an account you already have, or a business checking you for something that is not a credit decision.

The visibility rules differ too, and they matter more than most people realize. A hard inquiry appears on the version of your report that lenders see. A soft inquiry appears only on the version you see when you pull your own file. So a soft pull is not merely unscored, it is effectively invisible to the next lender who looks. That is why a page full of soft inquiries on your own report is nothing to worry about even when the company names are unfamiliar.

The last difference is consent. A hard pull requires your permission, which you usually give inside the application itself, often in a checkbox you barely read. A soft pull can happen without your involvement in several legitimate situations, which is why an existing card issuer can review your account or a firm can screen you for a preapproved offer without asking each time.

Who is allowed to pull your credit

Access to a credit file is not open to anyone curious. A company generally needs a permissible reason tied to a real transaction or relationship: you applied for credit, you applied for insurance, you are being considered for an account it already has with you, you applied to rent a home, or you gave written permission for an employment screening. The rules governing this sit in federal consumer credit law, and the details change over time, so treat the list here as the shape of the thing rather than a legal citation and confirm specifics with an official source.

What that means practically is that a hard inquiry on your report should always be traceable to something you did. You submitted an application, you signed a lease application, you asked a dealer to run financing, you opened a utility account at a new address. When you cannot trace it, that is meaningful, and the section on unauthorized inquiries later in this playbook covers exactly what to do.

It also means the permission you give is narrower than it feels. Authorizing one lender to pull your file does not authorize its partners to do the same. A single application that produces four hard inquiries from four different companies is worth questioning, even in situations like auto financing where a dealer shopping your application around is common and often disclosed.

Soft inquiries: the checks that cost you nothing

Concrete examples make the category obvious. Checking your own score in a banking app is a soft pull. Pulling your own full report from a bureau is a soft pull. A card issuer reviewing your existing account to decide whether to raise or cut your limit is a soft pull, sometimes called an account review. A lender screening a mailing list before sending you a preapproved offer is a soft pull. An insurance company quoting you a policy typically uses a soft pull, and in many states insurers use a credit-based score that is separate from the lending score anyway.

Prequalification tools sit here as well, which is the single most useful fact in this section. When a lender or a comparison page offers to show you likely terms without affecting your score, it is running a soft pull against your file, matching you to its published criteria, and returning an indicative answer. Nothing lands on the lender-visible report. You can do that as many times as you like.

The practical takeaway is that a great deal of credit shopping is genuinely free. You can see estimated rates, check whether you are in range for a card, and monitor your own score weekly, all without a single hard mark. The cost only arrives at the formal application. Our playbook on choosing a credit card leans on exactly that gap: do the comparison work in the soft-pull zone, and apply once you have decided.

Hard inquiries: the checks that leave a mark

The hard list is shorter and more predictable. Applying for a credit card creates one. Applying for a personal loan, an auto loan, a mortgage, or a student loan creates one. Requesting a credit limit increase on an existing card sometimes creates one, depending on the issuer, and this is the case most people are caught out by. Applying to rent an apartment often creates one. Opening a utility or mobile account in your own name sometimes creates one. A store card offered at a register creates one, at the least convenient possible moment for thinking it through.

Notice the common thread: in every case you asked a company to make a decision about extending you something, and the company needed your file to decide. The store-card case is the same transaction as a mortgage application in miniature, which is why it produces the same kind of mark.

The limit-increase case deserves a flag because it can surprise people who are doing the right thing. Raising a limit is one of the cleaner ways to lower reported utilization without paying anything down, a lever covered in our playbook on how credit utilization works. Some issuers grant increases from an internal soft review; others treat the request as a new credit application and pull hard. The way to find out is to ask before submitting, since most issuers will tell you which one they do.

How much a hard inquiry actually moves your score

Here is the number that dissolves most of the anxiety. A single hard inquiry on an established credit file commonly costs a few points, and an illustrative five points is a fair representation of the typical case. On a thin or young file, one with a couple of accounts and not much history, the same inquiry commonly bites harder, closer to an illustrative nine points, because there is less material around it for the model to weigh.

Multiple inquiries do not simply multiply. Models tend to show diminishing sensitivity as the count rises, so a burst of applications costs more in total than one but less than the count times the single-inquiry figure. The chart below shows the shape using illustrative figures for an established file, and every bar is drawn to its own value.

Read the chart against the rest of a credit report and the proportion becomes obvious. A run of six applications in three months, which would feel reckless to most people, lands an illustrative twenty three points. One card reporting near its limit on statement day can cost roughly twice that, and reverse it within a cycle. A payment reported thirty days late can cost several times more and take years to fade. Inquiries are real, but they are the small change of credit scoring.

How long a hard inquiry stays on your report

Two clocks run on every hard inquiry, and confusing them is the source of most of the misinformation on this subject. The first clock is visibility: how long the entry remains printed on your bureau report where a human underwriter can read it. That is commonly about two years. The second clock is scoring: how long the widely used models actually count the inquiry when calculating a number. That is commonly about twelve months.

The gap between the two clocks explains something people find strange. An inquiry from eighteen months ago is sitting on your report in plain sight, yet removing it would not change your score by a point, because the model stopped counting it six months earlier. A manual underwriter reviewing a borderline file might still notice a cluster of old inquiries and ask about them, which is a human judgment rather than a scoring effect.

The fade inside the scoring year is gradual, not a cliff. An inquiry from last week carries its full illustrative weight; the same inquiry at month nine carries a fraction of it. Nothing happens on a particular date that you would notice in your score. This is the opposite of how a late payment behaves, and it is why patience genuinely is the entire strategy for inquiry damage.

Visible for two years, scored for about one

It is worth spending a moment on how that split looks in practice, because it changes what you should actually worry about. Picture the twenty four months an inquiry occupies as a single bar. The first stretch is where the mark is at full strength and where a score dip is measurable if you are watching closely. The middle stretch is where the weight is bleeding away and where a score usually looks like it has recovered even though the entry has not moved. The final stretch, the whole second year, is scoring-irrelevant and matters only to a person reading the report directly.

That layout has a useful consequence for anyone timing a big application. If you have a mortgage in mind, the inquiries worth thinking about are the ones from the last six months, with a lighter concern about months six to twelve. Anything older is background. Trying to arrange your life around an inquiry from twenty months ago is effort spent on a number that has already stopped mattering, and the same effort put into what reports on your statement dates would do far more, as our playbook on raising a credit score sets out.

Rate shopping: why multiple mortgage or auto pulls count once

The scoring models contain a deliberate accommodation for the fact that shopping for a large loan responsibly requires several lenders to look at your file. Without it, comparing five mortgage lenders would cost five times what applying to one costs, which would punish exactly the behavior that saves borrowers the most money. So the widely used models group same-type pulls that land inside a short window and count them as a single event.

The window length varies by model version, and this is where honesty matters more than a tidy number. It is commonly cited as somewhere between fourteen and forty five days depending on which model a given lender runs, and lenders in different product lines routinely run different versions. Since you have no way of knowing which version will be used when your file is scored, the safe approach is to compress your shopping into the shortest window that still lets you compare properly. Two weeks clears every commonly cited version.

Many newer model versions also ignore mortgage, auto, and student loan inquiries from roughly the most recent thirty days when scoring, which gives a shopper a buffer before the pulls register at all. Treat that as a commonly described behavior rather than a guarantee, because the rules differ by model generation. The consistent, safe reading is simple: shop hard, shop fast, and finish.

Why the shopping window does not cover credit cards

The grouping applies to loan types where comparison shopping for a single purchase is the normal behavior: mortgages, auto loans, and student loans. It does not generally extend to credit cards. Apply for three cards in a week and you will usually collect three separate hard inquiries, because from the model’s point of view you were not comparing one product, you were opening three lines of credit.

That asymmetry is logical once you see it from the lender’s chair. Five mortgage pulls almost certainly end in one mortgage. Three card applications can easily end in three cards, three new limits, and three new accounts with no history. The first pattern is a shopper; the second is an expansion of borrowing capacity, and the model treats them differently on purpose.

The practical rule that falls out of this is worth memorizing. Compress loan shopping and space card applications. If you want two new cards, there is rarely a reason to apply for both in the same month, and putting several months between them lets the first inquiry start fading and the first account start aging before the second lands. How many cards to hold in the first place is its own question, handled in our playbook on how many credit cards you should have.

Prequalification vs preapproval vs application

The vocabulary here is genuinely confusing because lenders use the same words to mean different things. What matters is not the label but the type of pull behind it, so learn to look for that instead.

Prequalification is almost always a soft pull. You enter some basic information, the lender checks your file lightly against its criteria, and you get an indication of the terms you would likely receive. Nothing lender-visible is recorded. Preapproval is the ambiguous middle: in card marketing it often means the same soft process, while in mortgage lending a full preapproval usually involves a hard pull plus verified income and asset documentation, because it is meant to be something a seller can rely on. A formal application is always a hard pull, since a real decision is being made.

The test that works everywhere is to read the disclosure on the page in front of you before you submit. Reputable lenders state plainly whether the check is soft or hard, usually right beside the submit button. When the wording is vague, ask, because the answer is not something a lender has any reason to hide. The gap between an indicative soft answer and a binding hard one is exactly where careful shoppers do their work.

Why a card application and a loan preapproval behave differently

The two feel similar from the borrower’s side, since both end in a yes or no about credit. They behave differently because the underlying transactions are different, and knowing why keeps you from treating them as interchangeable.

A card application is a single, terminal event. You apply, one lender pulls, and either a revolving line opens or it does not. There is nothing to compare across lenders after the fact, because each card is its own product with its own terms. So the inquiry stands alone, and the score treats it as one discrete request for new credit.

A loan preapproval, especially in mortgage lending, is a stage in a longer process that is expected to involve several lenders. The pull is hard because the lender is doing real underwriting work, but the models anticipate that you will repeat the exercise elsewhere and group the results. The consequence for planning is that a mortgage preapproval is not a step to fear individually; the thing to manage is the calendar around it, so that every lender you talk to lands inside the same window. Unrelated card applications during that period are the genuine mistake, because they are neither grouped nor forgiven.

Applying for several cards at once: the stacking problem

The rebate-chasing habit of opening several cards in a short period does more than stack inquiries, and the inquiries are usually the least of it. Each approval opens a new account with zero months of history, which drags down the average age of your file. Each new account is a fresh line that a manual underwriter can see. And the applications themselves signal an appetite for credit at exactly the moment a big lender might be looking.

Put concretely with illustrative figures: four card applications in two months might cost an established file around seventeen points from the inquiries, then a further handful of points from the drop in average account age once the accounts open. The inquiry portion fades over the following year. The age portion takes longer, because average age only recovers at the speed of time passing.

None of this makes multiple cards a bad idea in principle. It makes clustered applications a bad idea when a mortgage, auto loan, or refinance is within a year. The same four cards opened one every six months would produce four inquiries that never overlap at full weight and four accounts that age in sequence rather than all at once. The cost of a card is mostly in its terms rather than its inquiry, which is the argument our playbook on choosing a credit card makes at length, and our debt payoff calculator shows what those terms turn into once a balance sits on the card.

How to see what inquiries are on your report

You cannot manage inquiries you have not looked at, and looking is free and soft. Pull your full report from each of the three major bureaus rather than relying on a score display, because inquiries live on the report, not on the score, and a company that pulled one bureau may not appear on the others.

Once you have the report open, find the inquiries section. It is usually its own block, often near the end, and it is typically split in two: inquiries shared with others, which are the hard ones, and inquiries shown only to you, which are the soft ones. That labeling varies in wording between bureaus, but the split is consistent, and the section headings on your own copy will make clear which list is which. Our playbook on reading a credit report walks the whole document section by section if the layout is unfamiliar.

Do this before any significant application rather than after. Checking a month before a mortgage gives you time to resolve anything odd, while checking after the decision leaves you arguing about something already priced in. It is also the cheapest fraud detection available, since an unexplained hard inquiry is often the first visible sign that someone is applying for credit in your name.

Reading an inquiry entry line by line

An inquiry entry is short, which is good news for anyone auditing one. You will generally see the name of the company that pulled the file, the date of the pull, and sometimes a category or an industry code indicating what kind of credit was involved.

Work through the list with your own memory as the reference. For each hard entry, ask whether you can place it: did you apply for that card in March, did that dealer run financing, is that unfamiliar name the servicing arm of a lender you did apply to. That last case accounts for a large share of the entries that look wrong at first glance, because the pulling entity is frequently a parent company, a bank behind a store brand, or a processor whose name never appeared on the application you filled out. A quick search on the company name usually resolves it.

Two patterns deserve a closer look. The first is a single application that produced several hard entries from different companies on the same day, common in auto financing where a dealer shops your application to multiple banks. That is often disclosed and often grouped by the models as loan shopping, but it is worth knowing it happened. The second is a hard entry you cannot connect to anything at all, which moves you into the next section.

Inquiries you did not authorize: what to do

An inquiry with no application behind it is either an error or a sign that someone is using your information. Both are worth acting on, and the process is the same at the start.

Begin by contacting the company that made the pull and asking what authorization it holds. Sometimes the answer is mundane, a mistyped identifier that attached its pull to your file, and the company can withdraw it. If the company cannot show authorization or does not respond, dispute the entry with the bureau reporting it. An inquiry that cannot be tied to a permissible purpose does not belong on your report, and the dispute route is the mechanism for removing it. The step by step process is laid out in our playbook on disputing a credit report error.

If the pattern suggests fraud rather than a clerical mistake, especially several unfamiliar inquiries in a short period, the stronger response is to freeze your credit at all three bureaus, which blocks new accounts from being opened in your name while the freeze is in place. Our playbook on freezing your credit covers how to place and lift one. Consider also that the inquiry may be the earliest visible symptom, and the accounts may not have appeared yet, which is why acting on an unexplained pull quickly is worth the hour it costs.

Inquiries you simply regret: the honest answer

Now the case nobody wants to hear about. You applied, you got the inquiry, and you wish you had not. Maybe you were declined. Maybe you changed your mind. Maybe you applied for a card at a register for a discount you no longer care about. The entry is accurate: you did apply, and the company did have your permission.

Accurate information does not come off a credit report on request. A dispute is a mechanism for correcting what is wrong, not for editing what is inconvenient, and filing one against an inquiry you know you authorized wastes the bureau’s time and yours. What you can do is nothing, on purpose, and it works: the entry stops counting toward your score after roughly a year, and it disappears entirely after roughly two.

There is one narrow exception worth mentioning honestly, which is that a lender will sometimes withdraw a pull it agrees should not have happened, for instance when a representative ran an application you had not agreed to submit. That is a conversation with the lender, not a dispute, and it depends on the lender agreeing that the pull was unauthorized. Outside that case, the honest answer to a regretted inquiry is that it will cost you a handful of illustrative points, fade within months, and be gone before it ever mattered.

Why paid inquiry removal services cannot deliver

This deserves saying plainly, because it is one of the most common pitches in the credit repair market. No company can remove a legitimate hard inquiry from your credit report. Not for a fee, not with a special letter, not through a contact at a bureau. The inquiry is accurate information about something you did, and the mechanism that removes inaccurate information does not apply to it.

What these services can actually do is file disputes, which is something you can do yourself for free, and then take credit for the entries that come off because they were genuinely unverifiable or because they aged off on schedule anyway. A service that disputes every inquiry on your report indiscriminately is also making assertions on your behalf that you know to be false, which is its own problem.

Weigh it against the size of the prize. Even in an extreme case, the total illustrative drag from a heavy application period runs in the low twenties of points, fading to nothing within a year. Paying several hundred dollars to chase that, in a market where the practical answer is to wait, is a poor trade. If the underlying problem is that your score is not where you need it, the money and effort belong in what reports on your statements and in your payment record, which is where the weight genuinely sits.

Where inquiries rank against the factors that actually matter

Step back to the whole scoring recipe and the proportion settles the argument. The commonly cited breakdown puts payment history at roughly thirty five percent of the calculation and amounts owed, which is essentially utilization, at roughly thirty percent. Length of credit history sits near fifteen percent, and credit mix and new credit take around ten percent each. Inquiries do not have a bucket of their own. They live inside the new credit bucket, sharing it with recently opened accounts, which means they are one component of the smallest slice.

Compare that with the leverage available elsewhere. Utilization responds within a single billing cycle, in either direction, and a card reporting near its limit can cost multiples of what an entire application spree costs. Payment history is heavier still and moves slowly in both directions. Between them those two buckets carry roughly two thirds of the calculation, which is why our playbook on why a credit score goes down starts the diagnosis there rather than with inquiries.

The conclusion is not that inquiries are irrelevant. It is that they are a rounding error compared with the things people worry about less. Someone agonizing over a single hard pull while carrying a card at eighty percent of its limit has the priorities exactly inverted. Pay down what reports on statement day, keep the payment record spotless, and the inquiry question becomes a matter of scheduling rather than damage control. What that paydown costs and how fast it works is what our debt payoff calculator is for.

How to plan applications around a big loan

Put all of it together into a calendar and the strategy is straightforward. Start with the date of the application that matters most, typically a mortgage or an auto loan, and work backward.

In the twelve months before it, apply for unrelated credit sparingly, since inquiries from this period are still being counted. In the six months before it, treat new card applications as something you need a real reason for, because these inquiries carry the most weight and any new account will be at its youngest. In the month or two before it, pull your own reports, resolve anything unexplained, and get your reported balances down so the file looks its best at the moment it is read. Then shop the loan itself hard and fast, gathering every lender pull inside a two week window.

Afterward, wait until the loan closes before applying for anything else. Lenders commonly re-pull credit shortly before closing, and a new card application in that gap has derailed transactions that were otherwise finished. The whole plan costs nothing but sequencing, which is the recurring theme of this playbook: the inquiry itself is small, and almost all of the available benefit comes from when you let it happen rather than whether you let it happen.

Common myths about hard inquiries

A few beliefs about inquiries are widespread and wrong, and they cause real harm.

The first is that checking your own credit hurts your score. It does not, ever. This myth is genuinely costly because it keeps people from looking at their own reports, which is where errors and fraud are found. Check weekly if you want to.

The second is that a declined application hurts more than an approved one. The score sees the inquiry, not the outcome. A decline and an approval produce the same inquiry, and the decline itself is not reported. What differs is what happens next: an approval adds a new account that affects age and utilization, while a decline leaves only the inquiry behind.

The third is that inquiries are the reason a score is stuck. Almost always the cause is utilization or a thin file rather than a handful of old pulls, and that diagnosis is the subject of our playbook on how long it takes to build credit. The fourth is that closing an account removes its inquiry, which confuses two unrelated records: the inquiry is a separate entry with its own two year clock, and closing the account does nothing to it while potentially harming your utilization and average age.

A worked example: two borrowers, twelve months

Two illustrative borrowers start the year with the same 740 score and the same established files, and they end it in very different places for reasons that have nothing to do with how much credit they used.

The first shops a mortgage. She pulls quotes from five lenders across twelve days, all of them hard pulls for the same loan type. Because they fall inside a tight window, the models group them as a single event. Her illustrative cost is around five points, taking her to 735, and by the time the loan closes she is back within a point or two of where she started. Had she spread those same five pulls across four months, they would have counted separately, at an illustrative twenty points, and she would have gone into her rate lock with a visibly weaker file.

The second applies for four cards over eight weeks chasing sign-up offers. Each is its own event, for an illustrative seventeen points from the inquiries alone, taking him to 723. All four are approved, so his average account age falls sharply, costing several illustrative points more. Nine months later he decides to buy a car. The inquiries have mostly faded by then, but his average age has not recovered, and the difference shows up in the rate tier he is quoted.

The lesson is not that one borrower was careless. It is that the same number of hard pulls produced roughly a four times difference in cost depending entirely on what they were for and how they were spaced. Timing is the whole variable.

The bottom line

A hard inquiry is a log entry that says you asked for credit, and it is the smallest ordinary mark a credit report carries. One commonly costs a few illustrative points, around five on an established file and closer to nine on a thin one, and it is scored for roughly twelve months while remaining visible for about twenty four. A soft inquiry, which covers every check you make on yourself and every look a company takes for marketing, account review, or prequalification, costs nothing at all and is not even shown to the next lender. Mortgage, auto, and student loan pulls of the same type inside a tight window are commonly grouped as one, so compressing loan shopping into two weeks is the single most valuable habit in this playbook. Card applications get no such grace, so space them out.

When an inquiry appears that you did not authorize, dispute it and consider a freeze, because it may be the first visible sign of something larger. When an inquiry appears that you did authorize and now regret, let it fade, and do not pay anyone who claims they can delete it, because nobody can remove accurate information. Then put the worry where it earns something: on what your statements report and on never missing a due date. Those two things carry roughly two thirds of the score. The inquiry carries a footnote, and footnotes expire.

What this playbook offers is general educational background on how credit inquiries are commonly recorded and weighed, and it is not financial, legal, or credit repair advice about your particular file. Every point figure, window length, and timeline in it is illustrative, chosen to show the shape of the mechanics rather than to predict what your report will do, and scoring model versions differ enough that outcomes vary meaningfully between borrowers. Bureau reporting practices, dispute procedures, and the rules governing who may access a credit file also change over time, so verify current details with the bureaus themselves or with the lender involved before acting. If an unexplained pull points to identity theft, or a large borrowing decision hangs on the answer, speak with a qualified professional, a nonprofit credit counselor, or an attorney rather than relying on any article, this one included.

Edited by Hamza Hai, MBA · Editor

Hamza Hai is the editor of BorrowLane. She holds an MBA and reviews the site's articles against our editorial standards, checking that every figure is labelled for what it is, that nothing is presented as verified fact without a source the reader can check, and that the writing stays useful to a non-specialist.

How we research, write and review · LinkedIn