Mobile Pre-Purchase Inspection: What a Mechanic Can Check Without a Lift

There is a particular kind of used car that cannot be inspected the normal way. It is parked outside somebody’s house four hours from yours, or it belongs to a seller who has already said no to letting a stranger drive it to a garage, or it is sitting on a lot in a state you have never visited and will not visit before the money moves. Every guide to buying a used car tells you to have it inspected first. None of them tells you what to do when the car cannot come to the inspection.

The answer is that the inspection goes to the car. A mobile inspector arrives in a van with a scan tool, a torch, a paint depth gauge, a trolley jack and stands, a tyre depth gauge and a camera, and works on whatever surface the car happens to be standing on. It is a real trade with real practitioners, and for a private-party sale it is frequently the only version of the step that is actually available to you.

It is also, unavoidably, a smaller examination than the one a workshop performs. Almost every page written about this subject says so and then stops, because saying so costs nothing and quantifying it costs work. This page does the work. We pulled the complaints that owners of twenty-seven mass-market nameplates actually filed with the federal regulator, sorted every one of them by the physical access an inspector would need to find that fault, and counted. What follows is what the lift is worth, in what proportion, and on which cars.

What the lift is actually worth, counted

Start with the shape of the problem. Any fault on a used car reveals itself by one of four routes, and each route needs a different thing from the person looking.

Some faults you find standing on its wheels: cracked wiper blades, a seat frame that rocks, a headlamp lens gone milky, a boot floor with water in it. Eyes, hands and a torch. Some you find only with the car off the ground: a corroded subframe seam, a split suspension bush, a brake disc you cannot measure until the wheel is off, a fuel line weeping where it runs over the axle. Some are invisible from every angle and surface only through the port, as a stored or historic fault code inside a control module that nobody ever cleared properly. And some present only under load, with the engine hot and the car moving — a gearbox that shunts on the two-three change, a misfire that only appears above a certain load, a cooling system that copes with an idle and not with a hill.

A mobile inspector can take three of those routes in full. They can look, they can plug in, and they can drive. What they cannot do without a workshop is raise the car properly. So the question worth answering is a countable one: of the faults that owners of ordinary used cars actually report, what share sits on the one route that a driveway closes?

To answer it we took NHTSA’s public owner-complaint database and pulled a fixed panel — the same twenty-seven nameplates, held constant, at three model years chosen to be twenty-one, eleven and three years old at the date of the pull. Eighty-one archived responses, 32,257 complaints, retrieved on 7 September 2026. Every complaint carries a component label written by NHTSA rather than by the owner, and we mapped that vocabulary once, in advance, to the four routes above. The mapping is published in full in the file behind this page, along with the four labels whose route is genuinely arguable, so anybody who disagrees with our judgement can move them and recompute.

Of the 32,257 complaints, 29,770 carried a component label that classified to one of the four routes. Another 2,460 were filed under unknown or other, which is a real category on NHTSA’s intake form and tells you nothing about where the fault lived; a further 27 turned out to be complaints about a child restraint product rather than about a vehicle. Both groups are reported separately and excluded from the denominator rather than quietly folded into it.

Against those 29,770, the split is this. Under load: 33.0%, or 9,829 complaints. Off the ground: 31.9%, or 9,495. Through the port: 23.6%, or 7,032. Standing on its wheels: 11.5%, or 3,414.

Read that as a buyer rather than as a statistician and it says something fairly blunt. Roughly two faults in three, on this measure, are reachable by somebody who never lifts the car at all, because they either announce themselves to a scan tool or they announce themselves on a road test. The remaining third is the price of the driveway. That is not a rounding error and it is not a catastrophe either; it is a third, and a third is exactly the kind of number that should be stated rather than gestured at.

The four routes, component by component

The mapping is the part that deserves scrutiny, so here it is with the counts attached. These are pooled across all three model years.

Four of those component labels are honestly contestable, and pretending otherwise would be worse than admitting it. Steering, at 3,067 complaints, is the biggest of them: racks, tie rods and ball joints all want the wheels off the ground, but an electric power-steering module that has failed would show on a scan instead, and on recent cars that is a large share of what gets filed. Fuel system, 959, has its tank, lines and pump under the car while its injector and pump faults usually announce themselves as codes. Wheels, 209, splits between kerb damage you can see standing up and bearing play you cannot. Engine and engine cooling, 1,321, we put under load, although a perished hose and a weeping water pump are a bonnet-up look.

So the file also carries a floor: the same share computed on nothing but the underside systems that nobody argues about — structure, suspension, service brakes, parking brake, trailer hitches. On that stricter measure the off-the-ground share is 17.7%, or 5,260 complaints. The true value of the lift sits somewhere between 17.7% and 31.9% of what owners report, and where in that band it sits depends on how you resolve four labels. We have given you both ends rather than one number with the argument hidden inside it.

One further check, because a complaint can name more than one system. 8,329 of the 32,257 do, which is 25.8% of the file. Our count assigns each complaint to the route of the first system NHTSA lists, so every complaint counts exactly once. If instead you count a complaint as needing the car raised whenever any system it names does, the total rises from 9,495 to 10,490. That is the ceiling on the measure, and it moves the result by less than the contested-component question does.

The older the car, the more the lift is worth

The pooled number is useful but it hides the finding that should actually change your decision. The panel was built with the same twenty-seven nameplates in all three model years precisely so that a shift between cohorts would be a shift in the cars rather than a shift in the sample. There is a shift, and it runs in one direction.

On the 2005 cohort — cars twenty-one years old, 14,373 classified complaints — the off-the-ground share is 37.0%. On the 2015 cohort, eleven years old, 11,842 classified complaints, it is 26.8%. On the 2023 cohort, three years old, 3,555 classified complaints, it is 28.2%.

The last two do not fall in a straight line, and the reason is worth stating rather than smoothing over: the 2023 cohort files a great deal against steering, 555 complaints or 13.7% of everything in that cohort’s file, and steering is the most contested label in the mapping. Take the contested labels out and use the strict underside floor instead, and the line is perfectly straight. Structure, suspension and brakes account for 20.6% of the classified complaints on the twenty-one-year-old cars, 16.0% on the eleven-year-old ones, and 11.3% on the three-year-old ones. The share of a car’s trouble that lives underneath it roughly halves across those two decades.

The individual components say the same thing more plainly. Structure — NHTSA’s label for the body and frame, which is where corrosion gets filed — is 1,222 complaints on the 2005 cohort, 8.3% of everything owners reported about those cars. On the 2015 cohort it is 828, or 6.2%. On the 2023 cohort it is 116, or 2.9%. Suspension runs 6.7%, then 2.5%, then 2.1%. Neither of those is a mystery: rust and worn bushes are functions of time and salt, and a three-year-old car has had neither in quantity.

Now look at the route running the other way. Electrical system is 12.9% of the twenty-one-year-old cohort’s file, 10.6% of the eleven-year-old one, and 18.8% of the three-year-old one — the single largest component on the newest cars in the panel. Forward collision avoidance does not appear at all on the 2005 cohort, appears 17 times on the 2015 cohort, and appears 119 times on the 2023 cohort, 2.9% of its file. The through-the-port route takes 24.7% of the oldest cohort, 20.4% of the middle one and 30.0% of the newest.

The practical instruction falls straight out of that. On a fifteen- or twenty-year-old car, a driveway inspection is giving up the route that carries the most risk, and you should either find a way to get it onto a ramp or accept that you are buying the underside unseen. On a three- or four-year-old car, the ramp matters less than the scan tool, and a mobile inspector with proper multi-module diagnostic equipment is close to a complete examination. If you take one thing from this page, take that the right answer depends on the car’s age and not on which format is generically better.

Two cautions on all of the above, because they matter. NHTSA complaints are filed voluntarily, so this is a description of what owners chose to report, not a failure rate; there is no fleet denominator anywhere in the source, so no per-vehicle rate can be computed from it. And the older cohorts have had far longer to accumulate filings than the newest one, which is exactly why nothing here compares raw counts between model years. Only shares within a cohort are compared, and only the ordering is safe to reason from.

What a mobile inspection does better than a workshop one

It would be a poor page that treated this as pure loss. There are three things the mobile format does better, and two of them are not obvious.

The first is the only one everybody mentions: it happens at all. A seller who will not release the keys, a car with no current registration, a private sale two states away, a vehicle already listed by somebody who works shifts and cannot spare a Tuesday morning — in every one of those cases the alternatives are not “mobile or workshop”, they are “mobile or nothing”. A partial examination beats no examination by an enormous margin, and the shares above are the size of that margin.

The second is that the car is cold. This one is genuinely undervalued. A car driven to a garage arrives warm, and a warm engine hides a great deal: the smoke on a cold start, the rattle from a timing chain tensioner in the first second before oil pressure builds, the misfire that clears once everything has expanded, the moment a diesel takes to catch. An inspector who arrives at the car gets the cold start, which a workshop appointment structurally cannot give them unless the seller agrees to be towed in. Given that a third of the complaints in our panel fall on the under-load route, the quality of the running examination is not a consolation prize.

The third is that the inspector sees the car in its habitat. Where a car actually lives tells you things a bay does not. Whether there is a stain on the drive, and what colour. Whether the tyres have flat-spotted from standing. Whether the interior smells of damp on a wet morning rather than after an hour under workshop lights. Whether the seller’s account of how the car is used matches the school run in the back seats and the roof bars on the roof. None of that goes in a report as a finding, and all of it shapes how the findings should be read.

What jack stands buy, and what they do not

Most competent mobile inspectors carry a trolley jack and axle stands, and will use them where the ground is level and hard enough. It is worth being precise about what that adds, because it is the single most oversold part of the format.

Stands under one corner at a time will let an inspector get a wheel off, look at the inboard face of a brake disc, measure pad thickness properly rather than squinting through an alloy, check a bearing for play, feel for a torn constant velocity joint boot, and rock a wheel to test ball joints and track rod ends. That is a real chunk of the off-the-ground route, and it covers a good deal of what “service brakes” and “wheels” mean in the table above.

What stands do not give is a walk down the length of the car. Corrosion assessment is not a spot check; it is a sequence, from the front subframe mountings along both sills and jacking points to the rear spring seats, the fuel and brake lines that run between them, the boot floor from underneath, and the seams where a repair after a collision would show. That job needs an inspector standing upright under the whole car with good light, and it is the reason structure is the component that separates a twenty-year-old car from a three-year-old one most sharply in our data. It is also the finding that most often ends a purchase, because unlike a worn bush it is not a job with a defined end.

An inspector who tells you plainly which of those two things they are offering is an inspector worth booking. One who says “we check underneath” without distinguishing them is describing a service and not a method.

Booking one: the questions that separate an inspection from a look

Because nobody licenses this work, the entire quality signal lives in the booking conversation. Ask these, in this order, and listen for specifics rather than reassurance.

What diagnostic equipment do you carry, and does it read every module or only the engine? This is the most important question on the list and the one least often asked. Nearly a quarter of the complaints in our panel name a system whose faults exist only inside a module. A code reader from a parts counter interrogates the engine controller and returns codes that are currently illuminating a lamp. That is not the same product. A proper scan walks the whole network — transmission, ABS, restraints, body, climate, driver assistance — and returns stored and historic faults as well as live ones, which is how you discover that the airbag warning was cleared last week rather than repaired. If the answer is vague, the through-the-port route is not covered and roughly a quarter of the risk in the table above is going unexamined.

Will you jack it, and what will you check with the wheels off? Covered above. The specific answer matters more than the yes.

Will you drive it, for how long, and on what roads? A third of the complaints in our panel fall on the under-load route. A lap of a residential street tests almost none of it. What you want is a run long enough to bring everything to temperature, including a stretch at speed and, if there is one nearby, a hill.

What do I get, and when? A written report itemising each system, the condition it was found in and the work each finding implies. Photographs, ideally with a scale reference on anything structural. The scan output as data, not as a sentence summarising it. And the report should belong to you whether or not you buy the car.

What can you not assess on this particular car, at this address? The best answer you can get to this question is a specific list, delivered before the appointment rather than after it. An inspector who says “on a car this age I would want it on a ramp for the sills and the rear spring seats, and I will tell you what I can see and what I cannot” is telling you exactly how to read their report. That sentence is worth more than any certification logo.

One more, and it is a screening question rather than a technical one. Who is paying you? An inspector introduced to you by the seller, or one whose fee is being covered by the seller as a courtesy, is not independent no matter how competent. Book your own, pay for it yourself, and have the report addressed to you.

The private sale is the one the federal rule does not reach

There is a reason mobile inspection exists as a trade at all, and it is not convenience. It is that the transactions in which the buyer has the least protection are also the ones in which the car is least likely to be brought to a workshop.

The Federal Trade Commission’s Used Motor Vehicle Trade Regulation Rule, 16 CFR Part 455, published at 49 FR 45725 on 19 November 1984, sets out duties for used-vehicle sellers. Section 455.1 makes it a deceptive act to misrepresent the mechanical condition of a used vehicle, to misrepresent the terms of a warranty, or to claim a car is sold with a warranty when it is not; it makes it an unfair act to fail to disclose before sale that a vehicle is sold without any warranty. Those are real obligations with a regulator behind them.

They attach to a dealer. And the same section defines the term: a dealer is any person or business that sells or offers a used vehicle for sale after selling or offering for sale five or more used vehicles in the previous twelve months, with carve-outs for banks, businesses selling to their own employees, and lessors selling to a lessee. Under the following definition in that section, a consumer is anybody who is not a used vehicle dealer. We read Part 455 from the eCFR on 7 September 2026 to confirm the wording.

So the person selling their own car in front of their own house is a consumer under the rule, not a dealer, and the federal prohibition on misrepresenting a vehicle’s mechanical condition is not the instrument that governs your conversation with them. State consumer-protection and fraud law may still apply, and this page is not the place to work out how in any particular state. The point is narrower and it is structural: the sale with the fewest federal guardrails is the sale most likely to require an inspector who will come to the kerb. The inspection is not a nicety in that setting. It is the substitute for a disclosure regime that does not apply.

It follows that the documentary side of the check matters at least as much here as the mechanical one. Decode the number on the car and confirm it is the model, engine and trim the advert claims. Check the open recall record against that specific VIN. Read the title and odometer history rather than the document in the seller’s hand. Our guide to checking a used car’s history before you buy sets out that sequence and what each federal source does and does not cover. Do it before you book anybody, because it is the cheap step and it eliminates cars.

Buying out of state, or without ever standing next to the car

The other case that forces the mobile format is distance. Cars are routinely bought across state lines now, from private sellers, from online retailers and from auction platforms, and delivered on a transporter to a driveway hundreds of miles from where they were photographed. A mobile inspector is the only way to put an independent pair of eyes on such a car before the money moves, and there are four things worth getting right.

Hire near the car, not near yourself. The inspector who is forty minutes from the vehicle will actually go; the one who is four hours away will quote for travel and may reschedule. Ask for photographs as a condition of the booking rather than as a favour, because you are going to conduct the entire negotiation without touching the car and a written finding without a picture attached to it is difficult to act on. Get the work done while the car is still standing on the seller’s premises, since a fault discovered after loading is a haulage problem as well as a mechanical one. And treat a return window as a genuine protection but not as a substitute for the inspection: the burden of noticing the fault falls on you, inside a short period, on a car you have just paid for and would rather like.

The age finding above bites hardest in exactly this situation. A three-year-old car bought at distance and inspected on a driveway with a proper module scan is a reasonably well-covered transaction. A seventeen-year-old car bought at distance and inspected on a driveway is a transaction in which the single largest category of risk — the underside — is being taken on trust. If that is the car you want, the right move is to pay the seller’s time or a transporter’s fee to get it onto a ramp somewhere local to it, rather than to convince yourself that the format difference is small. Our guide to what changes when the car is in another state covers the title and registration mechanics that follow.

What nobody publishes about mobile inspectors

This page has been specific about what it counted, so it should be equally specific about what it could not.

There is no federal or state licence for pre-purchase inspection. No agency registers the people who do it, publishes a count of them, or maintains a complaint file about their work in the way NHTSA maintains one about vehicles. Technician certification exists as a voluntary private credential and tells you that somebody passed a test in a subject area; it does not tell you that they carry a scan tool capable of reading a restraint module, or that they will write down what they could not reach.

Nor is there any published outcome data. Nobody publishes how often a mobile inspection finds a fault the buyer had not spotted, how often it kills a sale, or how it compares with a workshop inspection on the same vehicle. That study, if it existed, would be the most useful thing on this subject and we could not find a public version of it. The figures on this page are a proxy built from the other end: not what inspections find, but where on a car the faults owners report actually live. It is a good proxy and it is not the same measurement, and treating it as though it were would be exactly the sort of claim this page exists to avoid.

What that absence means in practice is that the burden falls entirely on you at the booking stage. There is no register to check, so the questions in the section above are the register.

How this fits with the rest of the check

A mobile inspection is one step in a sequence, and it is not the first one. The paperwork comes first because it is cheap and it eliminates cars before anybody has been paid to look at one. The inspection comes second because it is not cheap and it should only be spent on candidates that survived the record. Our full guide to what an independent pre-purchase inspection covers walks through the workshop version of the appointment in order, and our page on the three different things that get called a car inspection separates the state safety test from the buyer’s inspection from the selling dealer’s own free check — three products that share a name and prove very different amounts.

The one thing you should insist on yourself, before any inspector is involved, is that the car has not been moved on the morning of the appointment. A cold start is a diagnostic, and it is the only one in the whole process that a seller can quietly remove by being helpful. An inspector arriving at a car somebody has thoughtfully warmed up for them has lost exactly what the mobile format was supposed to buy.

Common questions

What is a mobile pre-purchase inspection?

An independent mechanic travelling to wherever a used car is parked and examining it there, on behalf of the buyer, before the sale. It covers the same ground as a workshop inspection — a visual and interior check, a full diagnostic scan, a road test, a paint depth reading, an identity and paperwork check, and a written report — with one exception. The car does not go on a ramp, so the underside examination is done with jack stands at best and from ground level at worst.

How much of a real inspection do you actually lose without a lift?

On our panel of 32,257 NHTSA complaints across twenty-seven nameplates, 31.9% of the 29,770 we could classify name a system that can only be reached with the car off the ground. Counted only on the underside systems nobody disputes — structure, suspension, brakes — the figure is 17.7%. So somewhere between roughly a sixth and roughly a third of what owners report is on the route a driveway closes. The rest divides between faults that surface on a scan and faults that only present on a road test, and a mobile inspector can take both of those routes in full.

Is a mobile inspection worth it on an older car?

It is worth it on any car, because the alternative in most of these situations is no inspection at all. But the gap between mobile and workshop widens sharply with age. On twenty-one-year-old cars in our panel, the underside systems account for 20.6% of classified complaints; on three-year-old cars, 11.3%. Corrosion of the body and frame alone falls from 8.3% of everything owners reported to 2.9%. On an old car, budget for getting it onto a ramp somewhere if you possibly can.

Can a mobile mechanic check for rust and accident damage?

Partly. A paint depth gauge works anywhere and will find filler and repainted panels without any lift at all, which covers a good deal of the accident question above the sills. Corrosion is the harder one: assessing it properly means following the sills, jacking points, subframe mountings, rear spring seats and floor seams end to end, and that needs an inspector standing under the car. Ask specifically what they will and will not be able to see on the surface the car is standing on.

Does a mobile inspector carry a proper diagnostic scanner?

Some do and some carry a basic code reader, and the difference is large. Nearly a quarter of the complaints in our panel — 7,032 of 29,770 — name a system whose faults exist only inside a control module: restraints, electrical, stability control, driver assistance. A parts-counter reader queries the engine controller and returns live codes only. Ask whether the tool reads every module and whether it returns stored and historic faults, not just current ones. It is the question that decides whether a quarter of the risk gets examined.

Will a seller let a mobile inspector come to the car?

Most will, and it removes the seller’s usual objection entirely, because the car never leaves their sight or their address. That is precisely why it is the accommodation to offer when a private seller refuses to let you take the car to a garage. If they decline this as well, you have your answer: a sale that is conditional on nobody qualified looking at the car is not a sale you want.

Is a mobile inspection enough for buying a car in another state?

It is usually the only option, and it is worth far more than nothing. Book somebody near the car rather than near you, insist on photographs alongside the written findings, and have the appointment happen while the car is still on the seller’s premises. Weight the documentary check more heavily than you would locally, since the recorded history is a larger share of everything you know when you cannot stand next to the vehicle.

Do mobile inspectors have to be licensed?

No. We could not find any federal or state licensing regime for pre-purchase inspection, any public register of practitioners, or any published data on what their reports find. Technician certification is a voluntary private credential. That absence is real and it is why the specific questions you ask when booking — what tool, what will come off, how long the drive, what you receive, and what they cannot assess at that address — carry the whole quality signal.

Does a mobile inspection replace a vehicle history report?

No, and the two answer different questions. An inspection of any kind tells you the condition of the car in front of you today. The record tells you what has been reported about its past — title brands, odometer readings, ownership count, which states it has been titled in — none of which any mechanic can see from underneath. Run the record first, because it is the cheaper step and it removes cars before you have paid anybody to look at one.

How was the data on this page produced?

We pulled NHTSA’s public owner-complaint database on 7 September 2026 for a fixed panel of twenty-seven nameplates at model years 2005, 2015 and 2023 — eighty-one archived responses, 32,257 complaints. Every complaint carries a component label written by NHTSA; we mapped that vocabulary once to four discovery routes and applied it mechanically, assigning each complaint to the route of the first system listed so that nothing is counted twice. The mapping is our editorial judgement, not NHTSA’s, and the four labels whose route is genuinely arguable are published alongside a stricter floor computed without them. Complaints are filed voluntarily, so none of this is a failure rate, a census, or a comparison between manufacturers.