Guide
How to decide which repair jobs to turn down
The jobs that damage a small repair business are rarely the ones that went wrong. They are the ones that should never have been accepted.
The short answer
Decide by category before you decide by job. Work out in advance which kinds of work you do not do, and the individual decisions stop being awkward: vehicles you are not equipped for, repairs that need a lift or a press you do not carry, anything that has to be done in a place you cannot work safely, and any job whose failure would put somebody in danger if it were done in a driveway. A category is a policy and it is easy to say out loud. A one off refusal sounds like a judgement about the customer.
For everything that survives the categories, run four checks. Can you do it properly with what you carry, in the place it has to happen. Does the price cover the whole day it will actually occupy, including the driving and the parts run. Does what the customer expects match what is really going to happen, including the possibility that the first repair is not the last one. And is anybody going to be able to tell whether it worked.
Turning work down when you are quiet feels wrong and is often the most valuable version of the decision, because a bad job taken in a slow week occupies exactly the days you needed for finding better work. The arithmetic to check that is your own: what the job pays, divided by the days it will genuinely take out of your week.
Decline on the work rather than on the person. Say what you do not do and why in one sentence, offer a direction if you have one, and stop there. Refusing service is governed by rules that vary by state and by federal law, and the notice at the end of this page says where to check rather than paraphrasing them.
Turning work away feels like losing and mostly is not
A booked day looks like a good day. That instinct comes from employment, where a full diary and a paid week are the same thing, and it survives into self employment where they are not. When you own the business, a day sold cheaply to a difficult job is a day removed from your only genuinely scarce resource, which is your own time.
The reason this is hard to see is that the cost of a bad job arrives later than the money does. The payment lands this week and the comeback, the argument, the unanswered messages and the review land over the following month, by which point they feel like separate events rather than consequences of one decision.
It is worth noticing which way your own error runs. Some people take everything and are permanently behind on the jobs they wanted; some refuse too readily and are permanently quiet. The fix is different in each direction, and knowing which one you are is more useful than any list of warning signs.
Decide the categories once, so the individual jobs are easy
Almost all the discomfort in declining work comes from making the decision in the moment, with a person waiting on the phone. The way out is to make the decisions in advance and in writing, when nobody is waiting and nothing is at stake.
Useful categories tend to be about equipment, environment and risk rather than about difficulty. Work that needs a lift, a press, an alignment rig or a controlled environment is not mobile work regardless of how competent you are. Work on a road, in an unlit car park, in an apartment garage with no space at either side, or in weather that stops you doing it properly, is a place problem rather than a job problem. And any repair where a failure would be dangerous and cannot be verified in the conditions available is its own category.
Vehicle types belong here too. Deciding you do not take on certain makes, ages or systems is a specialisation rather than a limitation, and telling a customer that you do not work on that particular thing is one of the easiest sentences in the trade to say.
The categories are also worth writing where customers can see them. A short line about what you do not do, on a booking page or in your first reply, filters a proportion of these before anybody has spent time on them.
Saying no without making an enemy
A decline is a small piece of customer service and it is worth doing well, because the person you turn down today talks to the same people as the person you take on tomorrow. The structure that works is short: what you do not do, one reason that is about the work, and a direction if you have one.
Giving a direction is the part that changes how the refusal is remembered. Naming a type of shop that handles it, or explaining what equipment the job needs, turns a rejection into help, and it costs one sentence. It also happens to be true, which matters more than the goodwill.
Two things to avoid. Do not invent a reason, because a fictional excuse is remembered as a lie the moment it is contradicted. And do not quote a deliberately high price hoping to be refused, because a proportion of people say yes, and then you are doing the job you did not want at a price you cannot defend.
The questions that decide it before you quote
These are the checks worth running on the booking call, before any time has been invested and while a no is still cheap. None of them is a rule and none of them predicts anything; they are the questions that surface the information a decision needs.
Can the work be done properly where it has to happen? Not whether it is possible, whether it can be done to a standard you would sign your name to. Space to work at both sides of the vehicle, ground that is level and solid, light, and somewhere for the fluids to go. A repair done badly because the location made it impossible is still your repair.
Do you carry what it needs? Specialist tooling, a diagnostic capability, a torque specification you can actually reach in that position. A job that is one missing tool away from finished is a second visit, and the second visit is usually unpaid.
Does the price cover the day rather than the hours? Work out what the job will really occupy: travel out and back, the parts run, the waiting, the setup and pack down. If that is most of a day and the invoice is worth an afternoon, the job is priced wrong even though the labour rate is right. This is your own arithmetic on your own figures and it settles more of these than any instinct does.
Is the diagnosis actually done? A great many bad jobs are bad because the fault was never identified, only guessed at, and the fitting turned out to be the easy part. If nobody has diagnosed it, the honest first job is the diagnosis, and that is a separate piece of work with its own price.
Will anybody be able to tell whether it worked? Intermittent faults, symptoms that only appear at temperature or at speed, and complaints described in terms nobody can measure are the jobs most likely to end in disagreement, because success has no definition. If you take one, define what a fix looks like in writing before you start.
Does what the customer expects match what will actually happen? Listen for the assumption that this is the whole problem, that it will be done today, and that the price cannot move. Where those assumptions are wrong, correcting them before the job is a conversation and correcting them afterwards is a complaint.
What happens if the first repair is not the last one? On an older vehicle with several failing systems, the job you are quoting is one of a series, and the customer may be about to spend more than the car is worth without knowing it. Saying that out loud is sometimes the most useful thing you do all week, and it occasionally costs you the job, which is the correct outcome.
Can you get paid? Not a judgement about the person, a practical question about the arrangement. Who is paying, when, and by what means. Fleet work with long terms, jobs where the payer is not the person present, and anything where payment depends on somebody else settling first are all workable, but they need to be agreed in advance rather than discovered afterwards.
What a bad job costs is never the job
The reason to run those checks is that the price of getting it wrong is not measured in the invoice. It is worth walking through where the cost actually lands, because the total is invisible if you only look at the job itself.
It costs the day, and then the days after it. A repair that goes wrong takes the original time, the return visit, the phone calls between them and the mental space that sits behind all of it. That last item is the one nobody counts and it is frequently the largest: a job you are worried about degrades the week around it.
It costs the other customers. Time spent on a job that should have been declined is time not spent on the ones that were fine, and they experience it as slower replies and later appointments. Those are the customers you actually wanted to keep.
It costs your judgement of your own prices. A job that turns into three visits reads afterwards as evidence that you are too expensive, because the customer paid once for what felt like a great deal of work. In fact it is evidence that the job should not have been taken, and the two conclusions lead in opposite directions.
It can cost publicly. A dispute that ends in a review outlives the money involved by years, and it is disproportionately likely to come from a job that had a warning sign nobody acted on at the booking stage.
And in the worst category it costs more than money. Safety related work done in conditions that do not allow it to be verified is the one place where the calculation is not commercial at all, and where the correct answer to a marginal case is always no.
Set against all of that, the cost of declining is one job you did not do. Written like that the trade is obvious, and the only reason it does not feel obvious in the moment is that the decline is certain and immediate while the damage is uncertain and distant.
Where the short answer stops being the answer
Every rule on this page has a situation it does not survive. Here are the ones worth knowing about before you meet them.
It is your quietest week and the phone is not ringing
This is when the checks matter most and when they are least likely to be applied. A marginal job taken in a quiet week occupies the exact days that finding better work would have needed, and it usually arrives with the reasoning that something is better than nothing.
Run the arithmetic rather than the feeling: what the job pays against the days it will really take. If the answer is poor, the alternative is not sitting idle, it is spending those days on the things that fill next month, and the guide in this cluster on getting customers covers what those are.
A good regular customer is asking for something you do not do
Say so, and say it as a fact about the work rather than as a refusal of them. Long standing customers generally understand the difference between somebody who will not help and somebody who is not equipped for a particular thing, particularly if you point them at who is.
Attempting it anyway to preserve the relationship is the version that damages it. A regular who watches you struggle with something outside your equipment loses more confidence than one who was told plainly on the phone.
The repair is straightforward but the location is not
Treat the place as part of the job, because for mobile work it is. No space at the sides, a slope, a surface that will not take a jack, no light, a car park that does not permit repairs, or a building management rule against it are all reasons the same repair is a different job here.
Where possible, offer the alternative rather than only the refusal: a different location the customer can get the car to, or a different day if the obstacle is weather. That converts a fair proportion of these into ordinary jobs.
Everything about the job is fine except the conversation
Pay attention to specific things rather than to a general feeling: a refusal to agree anything in writing, pressure to start before a price is settled, an account of previous mechanics in which everybody else was incompetent, or an insistence on what the fault is combined with a refusal to pay for a diagnosis.
Where you can, respond by tightening the arrangement instead of declining: written approval before work, payment on completion, the diagnosis as its own paid step. If those terms are refused, that refusal is the answer, and it is a decision about the terms rather than about the person.
When it has already gone wrong
Most people find a page like this after the fact rather than before it. This part is for them.
You are halfway into a job you should have turned down
Stop and re establish the agreement rather than pressing on quietly. Describe what you have found, what it now costs, what it will take, and what happens if it is not completed. Send it in writing and get an answer before doing anything else.
Where the honest answer is that the job needs equipment you do not have, say that too and say it early. Getting the vehicle to somewhere it can be finished is a bad afternoon; discovering the same thing at the end of the day with the car in pieces is much worse for everybody.
You accepted and now need to hand it back before starting
Do it immediately and by phone, then confirm in writing. The cost of withdrawing goes up every day the customer spends believing the job is arranged, and a car waiting on a booking that is not happening is the worst version of this.
Give a real reason and a direction, and where your withdrawal has actually cost them something, be generous about it. A well handled withdrawal is remembered far better than a badly done job.
You turned somebody down and they left a review about it
Reply once, briefly and without argument, saying what kind of work you do and that this fell outside it. That reply is for everybody reading afterwards rather than for the person who wrote it, and a calm answer to an angry review does more good than a defence.
Then check whether your public description of what you do is doing its job. Most complaints of this kind come from somebody discovering a boundary late, and the fix is to state the boundary earlier rather than to move it.
Questions
What jobs should a mobile mechanic not take?
The categories most people settle on are work that needs equipment a van cannot carry, work that cannot be done properly in the place it has to happen, and safety related work whose result cannot be verified in those conditions. Beyond the categories it becomes a judgement about the specific job, and the questions that settle it are on this page.
How do I politely turn down a repair job?
Say what you do not do, give one reason that is about the work, and point them somewhere if you can. Keep it short and do not invent an excuse, because a fictional reason is remembered as a lie as soon as it is contradicted. Quoting an inflated price to be refused is the other approach to avoid, since a proportion of people accept it.
Should I take work when I am slow even if the job looks bad?
Check it with your own arithmetic rather than with the feeling. Work out what the job pays against the days it will genuinely occupy, including travel and the likely return visit. A quiet week is exactly when a poor job does the most damage, because those are the days that would otherwise go into finding better work.
Can a mechanic refuse to work on a car?
What a business may and may not decline is governed by federal law and by state law, and it differs between states, so no web page can answer this for your situation. The practical guidance that keeps you on solid ground is to base a decline on facts about the work: the equipment it needs, the place it has to happen, the condition of the vehicle or the terms you were asked to accept. If you are unsure, ask somebody qualified in your state.
How do I know a job will turn into a comeback?
You cannot know, and this page assigns no odds to anything, because nobody here has measured it. What is worth noticing is whether success would even be recognisable: intermittent faults, symptoms that only appear in conditions you cannot reproduce, and complaints described in terms that cannot be measured are the jobs where nobody can agree afterwards whether the work succeeded. If you take one, define the fix in writing first.