Introduction — The Gears Were What They Noticed

Thirty-five minutes of competent work beside a brake that was going to fail.

The booking said "gears skipping, especially in the middle."

Bex arrived at eleven on a Tuesday, worked on the driveway with the bicycle in a portable stand, and found what the booking described. The rear cable was old, the housing was tired, and the indexing had drifted far enough that the chain was hunting between two sprockets under load. Bex replaced the inner cable, replaced the housing, cleaned and lubricated the chain, and set the indexing up properly rather than approximately — checked in both directions, under load, up and down the block.

Thirty-five minutes. The customer rode it up the road and came back grinning. It was, genuinely, good work. Bex was paid, wrote it up, and drove to the next job.

Eleven days later the same customer came down the hill on the way to the station, pulled the front brake, and it did almost nothing.

They were not badly hurt. They came off at the bottom where the road flattens, took the skin off one forearm and bent the front wheel, and were more shaken than injured. The bicycle went to a shop, and the shop rang them to say that the front brake pads had worn through to the metal backing plates and had been scoring the rim.

Not "were getting low." Worn through. Visible from a metre away, without tools, in about three seconds.

---

The half hour

Here is the part Bex has never quite got over.

For roughly thirty-five minutes on that Tuesday, Bex had been standing beside that bicycle. Bex had spun that front wheel. Bex had been at eye level with that brake calliper while sorting out the cable run at the front of the frame. Bex had looked directly at it and had not looked at it at all, because it was not what the job was.

The job was the gears. The customer had said gears. Bex had done gears, and had done them well.

⚠ Bex had fixed exactly what was asked for, competently and quickly, and had stood for half an hour beside a brake that was going to fail. The customer had not mentioned it because the customer did not know. That is what a customer is.

That last sentence is the whole book.

---

Why the customer could not have told you

There is an assumption buried in the way most repair work is booked, and it is wrong for bicycles in a specific and dangerous way.

The assumption is that the customer reports the problem and you fix the problem. That works when the fault announces itself. A car makes a noise. A boiler stops producing hot water. A mower will not start. The customer notices, the customer reports it, and the report is a useful starting point.

Now consider what the customer of a bicycle actually notices.

They notice gears that skip, because it happens every time they pedal hard and it is annoying. They notice a squeak. They notice a puncture, because the bicycle stops working entirely. They notice a saddle that has slipped, because they are sitting on it.

Now consider what they do not notice.

They do not notice a brake pad going from adequate to gone, because it happens over hundreds of stops, each one imperceptibly worse than the last, and the human hand compensates without ever reporting the compensation. They do not notice a rim sidewall wearing concave, because they cannot see the underside of a rim while riding and would not know what they were looking at. They do not notice a chain that has stretched past the point where it is quietly destroying a cassette, because a worn chain feels fine right up until it starts to slip. They do not notice a spoke that has gone slack, or a headset that has developed play, or a strand broken inside a cable housing, or a hairline crack under the paint of a fork crown.

⚠ The faults that are merely annoying announce themselves. The faults that injure people are silent.

This is not a failure of the customer. There is nothing a reasonable rider could do differently. It is a structural feature of the machine — and it means the customer's booking is, at best, a partial and unreliable account of what is wrong with the bicycle in front of you.

---

What that makes you

If the dangerous faults do not announce themselves, and the owner cannot detect them, then somebody has to look — and in the entire life of that bicycle between one service and the next, there is exactly one person who does.

You.

That is not a rhetorical flourish. Walk it through. Between the day you hand a bicycle back and the day it next sees a mechanic — which might be a year, might be three years, might be never — nobody will remove that wheel, nobody will look at the underside of that rim, nobody will measure that chain, nobody will squeeze that brake with a rag behind the pad to see how much is left. The rider will ride it. That is all that will happen to it.

So the state you hand it back in is the state it will be in, minus whatever the road does to it, for as long as it is ridden.

⚠ You are not being asked to fix a bicycle. You are being asked to decide that it is safe to ride away — and you are the only person in the chain who is in a position to make that decision.

Once you accept that, a great deal follows automatically. The inspection stops being a service you offer and becomes a thing you do. It stops being something you charge extra for and becomes something built into every price. It stops being skippable when you are running late, because "I was running late" is not a sentence you want to be saying afterwards.

And it means the customer's booking is where the job starts, not where it ends.

---

The second half of the problem

There is another thing that went wrong on that Tuesday, and it has nothing to do with brakes.

Bex did three jobs that day. They came to thirty-five dollars, forty-five dollars and thirty dollars. Call it a hundred and ten dollars for the day.

The work took thirty-five minutes, forty-five minutes and twenty-five minutes — one hour forty-five in total, which at a hundred and ten dollars is a shade under sixty-three dollars an hour. That is a respectable trade rate and it is the number Bex had in their head.

Now add the rest of the day. Twenty-two minutes to the first job. Thirty-one minutes to the second, because it was on the other side of town. Eighteen minutes to the third. Forty minutes home. Nine minutes total of parking, walking, getting the stand out and putting it away, three times over. Twelve minutes on the phone to a customer about a booking. Then, that evening, about twenty-five minutes writing the three jobs up and ordering two parts.

The day was not one hour forty-five. The day was four hours and fifty-two minutes, and it produced a hundred and ten dollars before fuel.

That is about twenty-two dollars an hour, and it is before Bex has taken anything out for tax, for the vehicle, for tools, for the parts consumed, or for the fact that some days there are two jobs rather than three.

⚠ Bex was not earning sixty-three dollars an hour. Bex was earning twenty-two dollars an hour and did not know, because nobody ever counts the driving.

This is the second structural fact of the trade, and it is why Rule 4 exists. The individual job in bicycle repair is small. The travel around it is not. A trade where the average job is worth forty dollars cannot absorb an hour of driving per job, and any business model that requires it will fail slowly and confusingly, while the person running it wonders why a decent hourly rate never turns into a decent week.

The solutions are all in this book — minimum call-outs, geographic days, clustering, workshop days, and above all recurring maintenance, which is the only thing that reliably converts a scattered map into a dense one. But none of them can be applied by somebody who has not first done the arithmetic honestly.

---

What this book does with those two facts

Everything else follows.

Because the dangerous faults are silent, there is a fixed inspection order that runs on every bicycle regardless of the booking, and it starts with brakes. Chapter 19 is that order. Chapters 20 to 24 are what you are looking at when you run it.

Because you are the last person who looks, there are things you will find that end the job rather than extend it — a rim past its wear limit, a fork with an unexplained history, a crack. Chapter 9 is that list, and Chapter 5.7 is how you tell somebody their bicycle is not safe to ride away when they are standing in front of you expecting it back.

Because a workstand is not a road, nothing goes back until it has been ridden. Chapter 25 is the protocol, and it is short, and it is not optional.

Because the job is small and the miles are not, the money chapters are built around working-hour cost rather than job price, and the growth chapters are built around density rather than volume. Chapter 8 does the arithmetic. Chapter 32 does something about it. Chapter 29 is the part that actually fixes it.

---

What it will not give you

It will not give you a single specification, and Chapter 37.3 explains at length why an AI system is a uniquely bad place to look for one.

It will not tell you what the law requires where you live, because it does not know and neither does anything you can ask.

It will not quote a price. Every figure is in U.S. dollars and every one is blank, because a printed number would be wrong for most readers and would be believed anyway.

And it will not promise you a living. Chapter 36.9 exists precisely because some people will do this properly and find the numbers do not work in their area, and the honest response to that is to stop, not to try harder.

---

Where to start

Not with tools, and not with a website.

Start with Chapter 14, which contains three items that stop work until they are resolved — one of which is whether your insurance covers a customer's bicycle while it is in your possession, which is a different question from whether it covers your work, and which most people discover the hard way.

Then Chapter 17, which is about competence, and specifically about finding the edge of yours on a bicycle that does not belong to a paying customer.

Then Chapter 19, and print the cards.

The first three jobs should take you twice as long as you think they will. That is correct. The order comes first and the speed comes later, and doing it the other way round produces somebody who is fast at missing things.

---

Bex still does that customer's bicycle. They came back, which was not a certainty, and they have been on a scheduled service ever since — twice a year, booked before Bex leaves.

The first thing Bex does now, on every bicycle, before the stand is even adjusted, is squeeze both brakes and look at what is left of the pads.

It takes about fifteen seconds.

---

⚠ What they brought it in for is not what is wrong with it.

---

©2026 James Henderson / https://localhandyman.work

Chapter 1 — Why Bicycle Repair Works as a Side Hustle

The only vehicle in common use with no scheduled maintenance requirement of any kind.

1.1 A trade that exists because almost nobody maintains a bicycle

There are a great many bicycles in the world and almost none of them are looked after.

This is not a moral observation. It is the commercial foundation of the entire trade, and it is worth understanding properly because it explains where your customers come from, what condition their machines will be in, and why the work is more available than the number of bicycle shops would suggest.

Consider what maintenance a car receives. It is registered, it is insured, in many places it is inspected annually by law, and the owner is prompted — by the dashboard, by the insurer, by the calendar, and by a legal requirement — to take it somewhere and have somebody look at it. Even an owner who cares nothing for cars ends up with the thing serviced, because the system will not let them avoid it.

Now consider what maintenance a bicycle receives. There is no registration. There is usually no insurance. There is no inspection requirement, no dashboard, no warning light, no annual prompt, and no authority that will ever ask. The bicycle sits in a shed or a hallway. When it works, it is ridden. When it stops working, it is put back in the shed.

The result is a population of machines that receive attention only at the point of total failure — and, critically, only for the specific thing that failed.

⚠ A bicycle is the only vehicle in common use with no scheduled maintenance requirement of any kind. Almost every one you touch will be overdue for something.

That is your market. It is not a market of enthusiasts who want their machine perfected. It is a much larger market of ordinary people whose bicycle has stopped doing something, who have no idea what maintenance a bicycle is supposed to receive, and who have never in their lives had one inspected.

What this means practically

It means the average bicycle that arrives in front of you will have several things wrong with it, of which the customer knows about one.

It means the phrase "when did it last have a service?" will very often be met with a genuinely blank look, because the customer did not know that was a thing.

It means you will spend a significant part of your working life explaining, kindly, that a bicycle needs looking after — and that this explanation, delivered well, is the single most valuable piece of marketing you will ever do, because it is what converts a one-off repair into a customer who comes back twice a year for a decade.

And it means the safety inspection at the centre of this book is not a service you invented to increase the bill. It is the thing nobody else in the chain has done.

1.2 What the customer is actually buying

They think they are buying a repair. They are not.

When somebody books a bicycle repair, the transaction in their head is simple: a thing is broken, you will fix the thing, they will pay you, the thing will work. That is the transaction they will describe to a friend, and it is the transaction they will judge you on if it goes wrong.

But the thing they are actually receiving is different, and understanding the difference changes how you price, how you talk, and what you write down.

What they receive is a bicycle that is safe to ride away, and a statement from a competent person that it is.

The repair is a component of that. It is not the whole of it, and it is not even the most important part. A customer who gets their gears fixed and then crashes has not received what they paid for, no matter how good the gear work was. A customer who gets their gears fixed and is told, honestly, that the rear tyre will need replacing before winter has received more than they paid for and does not yet know it.

Why this matters for pricing

If you believe you are selling repairs, you will price by the repair, you will feel awkward charging for an inspection nobody asked for, and you will end up doing the inspection unpaid or not doing it at all.

If you understand that you are selling a safe machine and a competent opinion, the inspection is not an extra. It is the product. The gear cable is a consumable that happened to be part of delivering it.

Chapter 27 builds pricing on exactly this basis, and Section 27.3 deals specifically with how to charge for an inspection that the customer did not request and does not yet value.

Why this matters for how you talk

Customers who think they are buying a repair get annoyed when you tell them about other things. It sounds like selling.

Customers who understand you have inspected the machine get reassured when you tell them about other things. It sounds like diligence.

The difference between those two reactions is entirely in how the work was framed at intake, before anything was touched — which is why Chapter 18 spends so long on the first four minutes.

1.3 Who needs this, and why they cannot do it themselves

The bicycle is often described as a simple machine that anyone can maintain. This is half true and the half that is false is the dangerous half.

Some bicycle maintenance genuinely is simple. Pumping a tyre, oiling a chain, adjusting a saddle height — these are within reach of almost anybody with ten minutes and a willingness to look something up.

The trouble is that the boundary between "simple" and "requires knowing what you are doing" is invisible from the outside, and it sits in a place that surprises people. Fitting a brake pad looks simple. Doing it in a way that does not leave a pad contacting a tyre sidewall, or a pad that will dive under the rim on a hard stop, is not simple, and the failure mode is severe.

So the honest answer to "why can they not do it themselves" has several parts.

They do not have the tools. Not the spanners — the measuring tools. A chain wear indicator, a torque wrench, a spoke key that fits. These are cheap and almost nobody owns them.

They do not know what to measure. Even with the tools, "is this chain worn" is a question with a specific answer that requires knowing the limit for that chain, which requires knowing what chain it is.

They do not know what they do not know. This is the real one. A person can watch a video, follow it exactly, and do a competent job of the thing in the video while remaining completely unaware of the three other things they should have looked at.

They have no reason to look. Nothing prompts them. See Section 1.1.

And a large proportion of them simply do not want to. This is entirely legitimate and it is most of your market. A person who cycles to work does not want a hobby. They want the bicycle to work and they will pay somebody to make that true.

The customer types this produces

The commuter. Rides five days a week in all weather, does high mileage, wears things out fast, and cannot afford to be without the bicycle. Values reliability and turnaround above almost everything. The best recurring customer in the trade.

The parent. Has between one and four bicycles in the garage, most of them small, most of them outgrown, all of them neglected. Comes to you in spring. Cares about safety in a way that is genuine and easily converted into a scheduled service, because they are already used to the idea that children's things need checking.

The returner. Bought a bicycle, rode it for a season, put it in a shed for three years, and has now decided to start again. Almost always brings you a machine with perished tyres, seized cables, a rusted chain and flat everything. Frequently the best-value job of the week, and frequently a bicycle that needs an honest conversation about whether it is worth it.

The leisure rider. Rides at weekends, in good weather, moderate mileage, moderate wear. Lower frequency, but often the easiest to convert to an annual service because the season gives a natural prompt.

The enthusiast. Knows a great deal, does most things themselves, comes to you for the things they cannot do or cannot be bothered to do. Small in number. Useful for referrals and dangerous for scope, because they will ask you to do things at the edge of your competence and will assume you can.

1.4 What it costs to start, honestly

Less than almost any comparable trade, and this is a genuine advantage.

A bicycle mechanic's minimum viable toolkit is a fraction of the cost of a motor mechanic's, a plumber's, or an electrician's. There is no expensive diagnostic equipment. There is no gas certification. There is no piece of capital equipment costing more than a modest amount. The single largest expense is usually the vehicle you already own.

Chapter 13 sets this out in three tiers with a worksheet, but the shape is:

Tier one — the things you genuinely cannot start without. Measuring tools, a workstand, a basic hand tool set, a torque tool, consumables, and insurance. This is a small number and it is the whole of what is required to do competent work on most bicycles.

Tier two — the things that make the work faster and safer. Better versions of tier one, specialist tools for the jobs you find yourself repeatedly turning away, a proper mobile setup, lighting.

Tier three — the things you buy only after the work has proved itself. Wheel building equipment, hydraulic service kit, a second stand, a van fit-out.

⚠ The commonest early mistake is buying tier three before tier one is complete, because tier three is more interesting.

The genuinely important cost is not a tool. It is insurance, and specifically the question of whether a policy covers a customer's bicycle while it is in your possession — a separate question from whether it covers your work, and one that Section 14.4 treats on its own because getting it wrong is the kind of mistake that ends a business rather than costing it money.

The other cost people forget is parts float. If you carry stock, you have money sitting in a box. Chapter 16 addresses how much and which.

Every figure in this book is in U.S. dollars and every figure is blank, because the numbers vary too much by location to print. Resource 5 is the budget worksheet.

1.5 The hours it fits into, and the hours it does not

Bicycle repair is unusually well shaped for a side hustle, and it is worth being specific about why.

It fits evenings and weekends. The customer base is mostly people who ride bicycles, and people who ride bicycles are frequently available at exactly the times a person with a day job is available. Saturday morning is the single busiest window in the trade.

The jobs are short. Most work is measured in tens of minutes rather than days. This means you can complete a job in an evening, which is not true of most trades.

The work is interruptible. A bicycle in a stand can be left in a stand. There is no setting material, no drying time, no half-finished thing that will be ruined by stopping. This is worth more than it sounds.

It has a natural low season. Chapter 32 treats this properly, but the shape is that late autumn through midwinter is quiet in most climates, and this is either a problem or a rest depending entirely on whether you planned for it.

The hours it does not fit

It does not fit unpredictable availability. If a customer books a Tuesday evening and you cancel twice, you do not have that customer any more. The trade runs on turning up.

It does not fit a schedule with no daylight. Mobile work in midwinter, at a kerbside, in the dark and the wet, is genuinely unpleasant and slower than you expect. Section 15.6 deals with it, but the honest answer is that a mobile-only business shrinks in winter unless you have somewhere to work.

It does not fit somebody who cannot say no. The trade offers you a constant stream of jobs that are slightly outside your competence, slightly too cheap, or slightly too far away. Chapter 9 is that list.

It does not fit anybody who wants to rush. This is the important one. The inspection takes the time it takes. A trade where the failure mode is a rider being injured is a trade with a floor under how fast you can go, and somebody who is temperamentally unable to accept that floor should not do this work.

1.6 What makes somebody good at it

Not manual dexterity, mostly. That is trainable and the bar is not high.

Order. The single strongest predictor of a good bicycle mechanic is whether they do things in the same sequence every time. The trade rewards process far more than talent, because the failures come from omission rather than from incompetence.

Curiosity about causes. A puncture is not a job until you know why it punctured. A cable that snapped did so somewhere, for a reason, and the reason is often still there. Mechanics who fix symptoms produce comebacks; mechanics who ask why produce customers.

The willingness to have an awkward conversation. You will regularly have to tell somebody that their bicycle needs more work than they expected, or is not safe, or is not worth repairing. This is the skill that separates people who last from people who do not, and it is a skill rather than a personality trait — Chapter 5.7 and Chapter 26 both teach it explicitly.

Comfort with saying "I don't know." In a trade with no specifications printed in this book and manufacturers who change things constantly, the correct answer is frequently "I need to look that up for your specific bicycle." A mechanic who cannot say that will guess, and guessing about a torque figure on a carbon steerer is how people get hurt.

Tolerance for the boring parts. Cleaning. Writing things up. Ordering parts. Chasing a supplier. These are most of the business and none of them are the reason anybody starts.

Physical care. Not strength. Care. The trade involves a lot of small precise things and one careless moment with a torque wrench can crack a frame that was fine.

1.7 The thing that makes this different from every other repair trade

Every trade has a customer who cannot diagnose their own problem. That is why trades exist.

What makes bicycle repair structurally different is the relationship between what the customer notices and what is dangerous — and specifically that the two are close to unrelated.

In most trades they correlate. A car that is unsafe usually gives some warning: a noise, a pull, a light, a smell. A boiler that is dangerous usually behaves oddly first. The customer's report is imperfect but it points roughly at the problem.

In bicycle repair the correlation collapses.

The things customers notice — skipping gears, a squeak, a slipping saddle, a slow puncture, a rubbing mudguard — are almost entirely things that will annoy them and never hurt them.

The things that hurt them — pads worn to backing plates, a rim worn concave, a fork with a crack, a stem bolt that was never torqued, a wheel that is not properly retained — are almost entirely silent.

⚠ In this trade the customer's report is not merely incomplete. It is systematically pointed away from the dangerous faults, because dangerous faults do not produce symptoms until they produce an accident.

That single fact generates every rule in this book. It is why the inspection is mandatory rather than offered. It is why it happens before the booked job rather than after. It is why the order is fixed. It is why you measure rather than judge. And it is why you cannot run this business on the assumption that a customer who did not mention brakes has brakes that are fine.

1.8 What you are risking every time you hand a bicycle back

It is worth being unsentimental about this, once, at the start.

When you hand a bicycle back you are making a statement. The statement is: this machine is fit to be ridden. You are making it whether or not you say it out loud, and the customer is relying on it whether or not they think about it.

The person who then rides it will do so in traffic, at speeds that are perfectly capable of causing serious injury, with no protective structure of any kind, in proximity to vehicles that weigh two tons. If the front brake fails at the wrong moment, or the fork fails, or the wheel comes out of the dropouts, the outcome is not an inconvenience.

This is not written to frighten anybody out of the trade. Bicycle repair is a good trade and the overwhelming majority of jobs are entirely routine. It is written because the standard of care in this work has to be set by the worst plausible outcome, not the typical one — and the worst plausible outcome is severe.

Practically, this means three things.

The inspection is not negotiable. Not when you are late. Not when it is a small job. Not when the customer is standing there. Not when it is a bicycle you did three months ago.

The record is not optional. If something does go wrong, the difference between a defensible position and an indefensible one is entirely whether you wrote down what you found and what you told them. Chapter 30 and Resource 31 exist for this.

Refusal has to be genuinely available to you. A mechanic who cannot afford to refuse a job will not refuse one. This is why the money chapters are safety chapters, and why Section 35.3 treats a year with zero refusals as a warning sign rather than a clean record.

1.9 Who should not do this work

Somebody who cannot afford to turn work away. See above. Financial pressure and safety judgement are directly opposed, and financial pressure wins.

Somebody who wants to work fast above all. There is a floor under the time. Speed comes from process, not from skipping steps, and it takes a year.

Somebody unwilling to keep records. The record is half the job. If writing things down feels like an imposition, this trade will punish that eventually and severely.

Somebody who cannot look something up. The absence of specifications in this book is deliberate, and it means the trade requires you to find the manufacturer's figure for the specific component in front of you, repeatedly, for years. Somebody who finds that tedious will start guessing.

Somebody who wants to avoid people. The awkward conversation is a core skill, not an occasional inconvenience.

Somebody who is not prepared to say no to an electric bicycle's battery and motor system. Section 9.8 draws that line precisely, and it is a line rather than a preference.

None of these are permanent disqualifications except perhaps the first, and the first is temporary. But it is better to know now than in month seven.

1.10 Chapter summary

Bicycle repair works as a side hustle because almost no bicycle receives scheduled maintenance, there is no authority that will ever prompt an owner to have one checked, and the result is a large population of machines that are overdue for something and owners who do not know it.

The customer believes they are buying a repair. What they are actually buying is a machine that is safe to ride away and a competent person's statement that it is — which is why the inspection is the product rather than an addition to it.

Startup cost is genuinely low by trade standards, and the largest and most easily-missed item is not a tool but the insurance question of whether a customer's bicycle is covered while in your possession.

The trade fits evenings and weekends well, is interruptible, and has a natural low season. It fits badly with unreliable availability, with a temperament that cannot accept a floor under how fast the work goes, and with anybody who cannot say no.

What makes it structurally different from every other repair trade is that the faults customers notice and the faults that injure them are close to unrelated. The annoying faults announce themselves; the dangerous ones are silent. That single fact produces every rule in this book.

---

Realistic scenario

Bex's fourth week. A Saturday morning, three jobs booked, and the first one is a returner — a bicycle out of a shed after four years, brought out because the owner's commute has changed and they want to try cycling in.

The booking says "needs new tyres, both flat."

The tyres are indeed flat, and both are perished, with the sidewalls cracked in the way tyres crack when they have sat for years partially deflated. That is a straightforward job and Bex has the tyres in the van.

Bex fits them, and then — because it is week four and Bex has read the order and is still slow enough to follow it properly — runs the inspection anyway.

The rear brake cable is seized inside its housing. Not stiff. Seized. The lever moves about a centimetre and returns halfway. It has been like that for some time and the owner has not ridden the bicycle in four years, so they have no idea.

The chain is beyond any measurement — surface rust bonded solid at four links, and it does not articulate at all in those places.

The headset has enough play to knock audibly when the front brake is applied and the bicycle rocked, which is a thing Bex now does on every machine because it takes two seconds.

Bex explains it. Not as a list of extras — as an account of what happened to a bicycle that sat in a shed for four years, which is a thing the owner can immediately understand and does not experience as selling.

The owner says the sentence Bex will hear a thousand more times: "I just thought it was the tyres."

They were right that it was the tyres. They were simply not in a position to know that it was also three other things, one of which meant the bicycle had effectively no rear brake.

The job goes from twenty minutes to an hour and forty. The owner approves it, because the explanation came before the work and because Bex showed them the lever rather than telling them about it. And Bex books them in for a service the following spring before leaving, which takes eleven seconds and is the most profitable eleven seconds of the morning.

⚠ The inspection took four minutes. Everything valuable about that job came out of those four minutes.

---

AI prompts for this chapter

Prompt 1 — Local Bicycle Population Map

"I am starting a bicycle repair side hustle. Help me build a picture of the likely bicycle population in my area so I can decide who to target. Ask me for my location type (urban, suburban, small town, rural), whether there is a rail station or major employer nearby, whether there are schools, cycle routes, or cycling clubs, and what the terrain is like. Then produce: (a) the likely mix of bicycle types in the area, as rough proportions with the reasoning for each; (b) which of the five customer types — commuter, parent, returner, leisure rider, enthusiast — are likely to be well represented and which are not; (c) three specific places or groups where those customers concentrate; and (d) what this suggests about which services I should lead with. Be explicit about which parts are inference from what I told you rather than fact, and tell me what I would need to check in person to confirm each one. Do not invent statistics or cite figures you cannot source."

Prompt 2 — Honest Self-Assessment Against the Trade

"Act as an experienced bicycle mechanic assessing whether I am suited to this trade. I will describe my current mechanical experience, my available hours, my financial situation, my tolerance for difficult conversations, and my record-keeping habits. Assess me against these specific requirements: working to a fixed process every time; curiosity about causes rather than symptoms; willingness to tell a customer their bicycle is unsafe; comfort saying 'I need to look that up'; tolerance for administrative work; and — most importantly — whether I am under enough financial pressure that I would find it hard to turn a job away. For each one, tell me honestly whether I am strong, adequate or at risk, and what specifically I should do about the weak ones before taking a paying job. Do not reassure me. If the financial pressure answer is concerning, say so plainly, because a mechanic who cannot afford to refuse work will not refuse it."

Prompt 3 — The Maintenance Gap Explanation

"Help me write a short explanation I can give customers, in plain conversational language, about why bicycles need scheduled maintenance when nobody ever tells them so. It must: compare it briefly to a car, which is inspected because the law and the insurer require it; explain that a bicycle has no equivalent prompt of any kind; explain that the faults an owner notices are the annoying ones while the faults that matter are silent; and end by making a scheduled check sound like ordinary sense rather than an upsell. It must not frighten anybody, must not imply their bicycle is currently dangerous, and must be sayable out loud in under sixty seconds without sounding rehearsed. Give me three versions: one for a commuter, one for a parent with children's bicycles, and one for somebody bringing a bicycle out of storage after years."

Prompt 4 — Available Hours and Realistic Capacity

"Help me work out how many bicycle repair jobs I can realistically take on per week alongside my existing commitments. Ask me: my available windows by day and their length; whether I intend to work mobile, from a workshop, or both; my likely travel time between jobs; and how long I currently take on a basic tune-up. Then produce an honest weekly capacity figure that separates working hours from travel hours from administrative hours, and shows all three. Assume I am new and therefore slower than I think — apply an explicit inexperience factor and tell me what factor you applied. Flag any window that is too short to be useful once travel and setup are counted, and any day where the travel exceeds the work. Present the result as a table plus three sentences of plain commentary."

---

⚠ AI checkpoint for this chapter

Before you act on anything an AI system tells you in this chapter, check these four things.

One — did it give you a specification? If any output above produced a torque figure, a wear limit, a tyre pressure or a service interval, discard that part entirely. This book contains no specifications and neither should any AI output you rely on. Section 37.3 explains why at length.

Two — did it invent a statistic? Market-sizing prompts are the single most common place where an AI system produces a confident, specific, entirely fictional number. If you were given a percentage, a market value or a participation rate, treat it as fiction unless you can trace it to a named source yourself.

Three — is the capacity figure honest? An AI system asked to calculate your capacity will tend to give you an encouraging answer. Check that travel time is counted in full, that administrative time exists at all, and that the inexperience factor was actually applied rather than mentioned.

Four — did it reassure you when you asked it not to? Prompt 2 asks for an unflattering assessment. If the output is warm and encouraging throughout, it did not do what was asked and you should push back explicitly.

---

Do This Now

1. Write down, in one sentence, what you believe you will be selling. Then rewrite it so that the word "safe" appears in it. Keep both. In six months, look at which one you actually behave as though you believe.

2. Count the bicycles you have access to that are not customers' — yours, family, friends, anything in a shed. That number is your practice fleet, and Chapter 17 will tell you what to do to them.

3. Start Resource 3, the dated source and rule register, today, with a single line in it. It will eventually be the most useful document you own and it only works if it was started early.

4. Answer honestly, on paper, the question from Prompt 2 about financial pressure. If the honest answer is that you could not comfortably turn away a job you were uneasy about, write down what would have to be true for that to change, and treat that as a prerequisite rather than an aspiration.

5. Ring or email one insurer this week and ask the Section 14.4 question — whether a customer's bicycle is covered while in your care, custody and control. Do not buy anything. Just find out what the answer sounds like.

---

©2026 James Henderson / https://localhandyman.work