How Much Does a Carfax Report Cost? What We Found, and When We Looked

You came here for a number, so here it is before anything else: when this site walked Carfax’s own purchase flow on 26 August 2026, a single report was priced at $49.99, two at $69.99 and four at $119.99. Now the part that matters more than the digits. On 29 August 2026 we went looking for that same figure on a page a member of the public can simply open — not a summary of one, not a range someone else had quoted, the company’s own price on its own site. It is not there. The number becomes visible only once you are inside the purchase flow, at the point where you have already supplied a VIN and are being asked for a card.

That is why our figures carry a date and why you should distrust any that do not, including ours once it has aged. A price that lives inside a checkout is a price nobody can re-read without starting again, so it decays into rumour the moment it is copied. The rest of this page therefore concentrates on the part that does not change every quarter: the structure of what you will be charged, why that structure exists, how to re-run our check in a minute if you want today’s number rather than August’s, and which of the money you are about to spend is genuinely necessary.

The number is not on a page you can reach

Say that carefully, because the careless version of it is wrong. Carfax has a price. It is not hiding a secret, it is not refusing to sell to you, and there is nothing improper about how it operates. What it does not do is put the retail figure on an open page, in the way a supermarket puts a price on a shelf edge. The number lives one step further in, past the point where you have committed some effort to the transaction.

This matters more than it sounds, because of what it does to everybody else’s writing on the subject. A search for the cost of a vehicle history report returns dozens of pages that all state a figure with total confidence, and many of them state the same figure, which reads like corroboration. It is not. Trace those pages back and the trail almost always ends at a comparison table published by a competing report provider — a company with an obvious interest in how expensive its rival looks, quoting a price it did not have to justify to anyone. That is one competitor describing another. It might even be right. There is no way for a reader to tell, and a number you cannot check is worth less than an honest blank.

So we did the only thing that produces a number a reader can weigh: we walked the flow to the point where the price appears, wrote down what it said and the day we saw it, and stopped there. That is what the figures at the top of this page are — one observation, dated, made by us. It is not a rate card, we cannot promise it survived the week, and we would apply exactly the same discount to a figure that flattered our own commercial interests.

Confirming it takes about a minute, and you should

You do not have to take our word for any of this, and the check is quick enough that it is worth doing on whatever provider you are considering rather than only on the one this page happens to be about.

The result of that check is the useful thing, and it is a real distinction between providers. A company that shows the price on its front page is one you can comparison-shop from an armchair. A company that shows it at the payment screen is one you can only comparison-shop by walking each competitor’s flow to the last step, which takes long enough that most people simply do not do it. Neither is dishonest. They are different commercial choices with different consequences for you, and knowing which you are dealing with is worth the sixty seconds.

Run it on Carfax and you will land where we landed: nothing on the product page, nothing in the help pages, and the figure appearing at the payment step. That is the whole basis for the numbers at the top of this article, and it is the reason we would rather teach you the procedure than ask you to trust a screenshot. If what you find differs from what we recorded on 26 August 2026, yours is the current one and ours is history.

Why a company would price this way, without assuming the worst

It would be easy, and cheap, to write a paragraph here implying that an unpublished price is a suspicious price. It is not, and the ordinary reasons are worth setting out properly, because a reader who understands them will make better decisions than one who has simply been told to be indignant.

Published prices are a commitment, and prices here move. Promotions run, bundles get repackaged, tiers are added and retired. A figure printed on a marketing page has to be updated everywhere it appears every time any of that happens, and the cost of getting it wrong — a customer arriving with a screenshot of an old number — falls on the support team. Keeping the price in one place, generated at the moment of sale, removes an entire category of that problem.

The consumer report may not be the main business. Vehicle history data is sold to dealer groups, marketplaces, lenders, insurers and fleet operators, generally on subscriptions negotiated privately. Against that, one-off retail sales to private buyers can be a modest slice of revenue. A company in that position has little reason to build its public pages around a retail sticker, and every reason to build them around the trade relationships that pay for the data in the first place.

A price out of context invites the wrong comparison. The products in this market genuinely are not equivalent, and a figure sitting next to a rival’s figure invites a reader to treat them as though they were. From the seller’s side, that is a real problem rather than an excuse: it is much easier to explain what you are buying inside a flow you control than in a table someone else built. How differently these products are assembled, and which records sit behind each one, is the subject of our guide to the alternatives to Carfax and what each of them is really selling — and it is the strongest argument against reading this whole subject as a price race.

And, plainly, comparison friction favours the seller. This is the part nobody in the industry says out loud, but it is not sinister and it is not unusual. Any business that makes it slightly harder to line its price up against a competitor’s will lose slightly fewer customers at that moment. Airlines do it. Insurers do it. Mobile networks have built entire tariff structures around it. It is legal, it is common, and recognising it is simply part of being a competent buyer.

Hold all four together and you get a fair reading. The pricing is opaque to a shopper, for reasons that are mostly mundane and partly commercial, and the practical consequence for you is only this: you cannot plan your budget from the outside, and you should not trust anyone who tells you they can plan it for you.

The shape of the offer, which is the part you can rely on

Strip out the numbers and the market has a consistent structure. Every consumer-facing provider in this business, large or small, sells through some combination of five shapes, and once you can name them you can work out what you are being offered even when the figures are hidden until checkout.

Notice that only two of the five rows describe you paying a list price for a report you chose. The other three are a taster, a trade product, and somebody else’s purchase. That is a reasonable description of how most private buyers actually end up holding a vehicle history report, and it is why the question “what does one cost” has a less useful answer than it seems to.

The free preview, and where its usefulness stops

A preview that asks for no payment details is doing a specific job: it proves the VIN resolves to a real vehicle and tells you roughly how much material exists behind the paywall. That is genuinely worth having before you spend anything, because a VIN that returns nothing at all is either mistyped or worth a question to the seller.

What a preview cannot do is settle anything. It will not give you the dated odometer sequence, the brand history or the ownership chain, because those are the product. Treat it as a check on the VIN and a look at the size of the file, not as a cheap version of the answer.

The subscription, which is the shape that catches people

Of the five, the recurring charge is the one that most often surprises a consumer, because it is the one where the amount you eventually pay bears no relation to the amount you thought you were agreeing to. The mechanism is boring: you wanted one report, the cheapest route to one report was a trial or an introductory month, and nothing then prompted you to cancel.

The defence is equally boring and works every time. At the moment you buy, find the words that describe what happens next month. If they say a renewal is coming, put the cancellation in your calendar for a few days before it, on the same day you buy. Not later. Later is how the second charge happens.

Bundle arithmetic, and why singles are the wrong purchase for a shopper

Multi-report bundles are close to universal in this market, and they are not a generosity. They exist because of a specific problem the seller has: a buyer comparing four cars will happily buy one report, look at it, and then decide that four reports is more than they want to spend. The bundle is designed to stop that decision happening, by making the second, third and fourth reports feel almost free at the moment you commit to the first.

Understanding the motive does not make the offer bad. For a buyer genuinely working through a shortlist, the bundle is straightforwardly the right purchase, and paying single-report rates on car after car is one of the more expensive mistakes available in a used-car search. The arithmetic is not subtle. It just requires you to be honest about a question most shoppers get wrong in the other direction.

How many cars will you actually pay to check? Not how many are on your list. Not how many tabs are open. How many will still be live after you have looked at the pictures properly, read the description twice, established that the seller responds to messages, and run the checks that cost nothing. For most people that number is much smaller than the list, and it is the number the bundle tier should be chosen against.

The scale of it is easy to underestimate. On the figures we recorded, the four-report tier works out at $30.00 a report against $49.99 bought singly — the fourth report costs you about what the first one did, and the two in between are close to free. Get the tier wrong in either direction and you have overpaid anyway. Buy singles when you were always going to check five cars and you have paid the highest available rate five times over. Buy the largest bundle for a search that was realistically two cars and you have bought credits you will never redeem, which is a discount on reports you did not need. The tier that fits is the one that matches the count you can defend, and if you are genuinely unsure, the smaller bundle plus a single later is usually cheaper than the reverse mistake.

Why “free Carfax” is everywhere, and who actually paid for it

Search for the cost of a report and half the results will be about getting one for nothing, which seems like a strange thing for a market to be full of. It is not strange at all once you see where the money is.

Dealers do not buy reports one at a time at whatever a private buyer pays. They hold trade accounts, priced on volume, and at that scale attaching a report to a listing costs a dealer very little per car. What it buys them is worth more than it costs: a listing with a report attached pre-empts the first question every used-car buyer turns up with, saves the salesperson an argument, and quietly says that nothing is being withheld. So the report on the car in front of you on a dealer’s lot is very often free, and it is free because the person selling you the car has already paid for it.

State the implication plainly, because it is the useful part, and then stop — it does not go as far as some people take it. A report that a seller bought and chose to publish was bought to answer the seller’s question, which is “how do I sell this car faster?”, and not yours, which is “what is wrong with this car?”. Those two questions overlap heavily on a clean vehicle, which is exactly why reports get published at all: a dealer volunteers one most readily on stock that has nothing awkward sitting in its record.

None of that makes the document less genuine. It is a real report with real records in it, and reading it costs you nothing, so read it. What it means is narrower and more practical: the fact that a report was free tells you something about the seller’s confidence and nothing about the car’s condition, and a seller who has clearly read a report and not offered it has communicated something without saying a word. The various legitimate routes to a report somebody else paid for — and the ones that are lead-generation dressed up as generosity — are set out in our guide to getting a Carfax report without paying for it.

There is a second document on a dealer’s car that costs you nothing and outranks any report the moment something goes wrong: the window form the Federal Trade Commission requires on used vehicles, which records the warranty position on that particular car and becomes part of what you eventually sign. People spend money on history reports and walk straight past that form, which is an odd allocation of attention given that one of them describes the past and the other decides who pays for the gearbox in November.

What you actually spend is set by the search, not by the report

Here is the reframing that saves people the most money, and it is why a page about price should not really be a page about one number.

Your total outlay on vehicle history data across a car search is the number of cars you pay to check, multiplied by the rate you are paying per report, plus whatever you spent on things that were already free. Three terms. Most buyers spend all their energy trying to shave the middle one, which is the term they control least, and ignore the first, which they control entirely.

The count, which is the term that matters

Every car you remove from the paid list before paying is a whole report saved, which is a far larger effect than any discount. The tools for removing cars cost nothing: the federal VIN decoder tells you whether the vehicle in the advert matches the vehicle the manufacturer built, and the federal recall lookup tells you what safety work is outstanding. Neither charges anything, neither needs an account, and between them they end a surprising number of conversations before a card comes out. Where each of them belongs in the running order is laid out in the full sequence for checking a car’s history.

The discipline is simple to state and hard to follow when you have found a car you like: nothing gets paid for until it has survived the checks that are free. A shortlist of eight becomes a paid list of two remarkably often, and the two are the ones worth spending real money on.

The rate, which is the bundle question

Once you know the count, the rate follows from it. This is the whole of the bundle decision described in the previous section, and it is worth doing in that order — count first, tier second — because doing it the other way round means letting a price ladder tell you how many cars you are shopping.

The waste, which is the term nobody audits

Three kinds of spending in this market buy you something you already had:

• Paying for a report on a car whose seller is holding one. Ask before you buy. On a dealer’s own stock this costs a message and saves a full report.
• Paying for recall information. Safety campaign data is published by the federal government at VIN level for nothing, and the remedy itself is performed at no charge by a franchised dealer. When an open campaign surfaces inside something you paid for, the finding is real and the record behind it was already public.
• Buying a second opinion on a car you have already decided against. This is rarer but it happens, usually when a first report turns up something ambiguous and the instinct is to check it elsewhere rather than to walk away or to pay a mechanic. If the finding was serious enough to worry you, a second database is the wrong next purchase.

And one caveat on the whole exercise in saving money, because it is the way this goes wrong most often: money you save on reports is only saved if it does not come out of the inspection budget. The reports read paperwork. An independent pre-purchase inspection reads the car, finds things that were never reported to anyone, and costs considerably more for exactly that reason. Trimming the cheap step to protect the expensive one is sound. Doing it the other way round is how people buy cars with immaculate records and a failing gearbox.

The figures on this page you can re-read at any time

Everything quoted so far had to be gone and fetched from inside a checkout, which is why it carries a date and a warning. One set of numbers on this page is different in kind: it is first-party, it sits on an open page anyone can load right now without a VIN or a card, and it can therefore be audited rather than trusted. It also belongs to the company that pays for this site, which you should know before you read it.

The practical difference is re-readability. We can open our partner’s pricing page, read what it says and write down when we read it, and so can you, in one click and with nothing typed in — which is precisely what checking the figure at the top of this page would cost you. The rates below were taken on 29 August 2026 and are recorded in the site’s own data file with that date attached.

A single report is $14.99. The per-report rate falls with volume in the ordinary way: $12.00 each in a three-report bundle, $11.00 each in a five-report bundle, and $9.00 each in a ten-report bundle, which is the tier aimed at people working through a long shortlist and at small trade buyers. Ahead of all of it there is a free preview that asks for no card, confirming the year, make, model and trim, showing real photographs of the vehicle where any are on file, and telling you how many recall, title and auction records exist before you decide whether the detail is worth buying. Payment is one-off rather than recurring, and there is a refund where a VIN returns no data at all.

Two honest caveats on that, because they apply to any price in this market. Prices change, so the figure that matters is the one on the provider’s page on the day you buy rather than the one in any article, including this one — you can check the current per-report rate at source in less time than it took to read this paragraph. And a cheaper report is not automatically a lesser one, nor automatically an equal one: what determines whether a report answers your question is which record sets sit behind it. Some of those are federal and compulsory — the consolidated title and salvage file that state agencies, insurers and recyclers are obliged to feed — and some are commercial feeds a provider had to go out and negotiate. That is a different subject from price, and we treat it separately in how accurate these reports actually are.

What we are deliberately not doing is building the obvious comparison table out of these two sets of numbers. They are not the same kind of fact. One is a live page either of us can reload; the other is a snapshot taken inside a checkout on a particular day, which nobody can re-open without starting a purchase. Setting them side by side in a tidy grid would imply both are equally current, and the arrangement would happen to favour the side that pays us. Read them as what they are, in the order they appear, and do your own check before you buy anything.

The terms that decide what you actually pay

The headline rate is the smallest part of what a report ends up costing you. Four terms sit underneath it, and all four are readable before you pay if you look.

One-off or recurring

The single most important distinction, and the one most easily missed at the moment of purchase because the screen is designed around completing the sale rather than around explaining the billing. Find the sentence that describes next month. If there is not one, look harder; if there is one and it mentions renewal, decide now whether you want that.

What happens when the VIN returns nothing

Occasionally you pay for a report on a vehicle with almost no reported history. Sometimes that is the answer — nothing was reported because nothing happened — and sometimes it means the VIN sits outside the provider’s coverage entirely. Providers handle this differently, and the terms are usually explicit about it. A provider that refunds where there is genuinely no data on a VIN is charging you for an answer rather than for a query, and that is a meaningful difference in what you are buying.

How many VINs one purchase covers

A report is normally tied to a single VIN and cannot be moved to another car if that deal falls through. This catches people who assume a purchase buys them a lookup rather than a document. If you are checking a car that might not still be for sale by the weekend, that is a reason to look at the preview first rather than a reason to hurry.

Whether you can get it back later

Reports are documents, and documents get lost. Some providers keep your purchases available in an account; some deliver once and that is that. It matters more than it sounds, because the useful moment for a report you bought in March is often August, when you are selling the car and would rather hand a buyer a record than an assurance. Save a copy at the moment of purchase regardless of what the provider promises.

Four ways buyers overpay, in rough order of how much they cost

Everything above condenses into a short list of expensive habits.

There is a fifth that belongs on a different page, so it gets one line here: buying a second report on the same car is rarely the best use of the next chunk of money, and the cases where it is defensible are narrow. We work through them in the head-to-head between the two big commercial reports.

Putting it in order

The sequence below is arranged so the free filters run first and nothing expensive is ever spent on a car that a cheap check would have eliminated. It is the whole money argument of this page in six lines.

Follow that and the money lands where it does work. Most cars leave the list during the free steps, the paid reports are bought on vehicles that are genuinely still candidates, and the largest single expense in the process is the one that can actually find a problem nobody ever reported.

Common questions

How much does a Carfax report cost?

When this site walked the purchase flow on 26 August 2026, a single report was $49.99, two were $69.99 and four were $119.99. That is an observation with a date on it rather than a rate card, because Carfax publishes no price on any page reachable without entering the flow — we looked again on 29 August 2026 and it is still not there. Figures quoted elsewhere have usually come from a rival provider’s comparison table rather than from the company, and prices here move with promotions and packaging, so if the number matters to your decision, walk the flow to the point where it appears and stop.

Other sites print the same figure without any of these caveats. Why the fuss?

Because a price with no date attached is how a wrong figure ends up circulating for years. Every number in every article on this site has to trace back to something we can point at, and ours points at a checkout screen on a named day. A price read off a competitor’s marketing page does not clear that bar, and it would not clear it if the number happened to make our own partner look good. The caveat is not modesty; it is the only thing that tells you when to stop believing us.

What does a vehicle history report cost in general, then?

The structure is more informative than any single figure. Consumer providers sell a free preview tier, a single report, and multi-report bundles where the per-report rate falls as the count rises; some also sell recurring subscriptions, which are aimed mainly at the trade. Our own partner’s published rates, disclosed above as a commercial relationship, run from $14.99 for a single report down to $9.00 each in a ten-report bundle, which gives you a sense of the spread between the single and bundled rates in this part of the market.

Are multi-report bundles genuinely cheaper, or just a way to sell more?

Both, and there is no contradiction. The per-report rate really does fall, and the bundle really is designed to stop you deciding after the first report that you have spent enough. It is the right purchase if you are honestly going to check several cars and the wrong one if you are not, since credits you never redeem are a discount on nothing. Count the cars you will actually pay for, then pick the tier.

Is the free report on a dealer listing a real report?

Yes. It contains the same records it would if you had bought it, and it costs you nothing, so read it. What it does not do is tell you anything about the car by virtue of being free — it is free because the dealer bought it as a sales tool, and dealers publish reports most readily on the cars where doing so helps them. Treat it as a genuine document produced for somebody else’s purpose.

What is the cheapest way to check several cars at once?

Run the free federal VIN decode and recall lookup on every car before spending anything, which typically removes most of the list. Ask the remaining sellers whether a report already exists. Then buy one bundle sized to whatever survived, rather than paying single rates repeatedly. The saving comes overwhelmingly from cutting the count, not from finding a cheaper provider.

Does a free VIN check turn into a paid one?

Sometimes, and the tell is when the page asks for a card. A genuine free preview returns something before any payment details are requested — usually a confirmation of the vehicle and a count of the records on file. A page that collects card details or an email address before showing you anything is selling you something, whatever the heading says, and it is worth knowing which of the two you are on before you type a VIN into it.