Repair or Replace?

Should I Repair or Replace It? The Complete Decision Framework

Your dishwasher has stopped heating. Your laptop won’t hold a charge. The fridge is making a noise it didn’t make last month. Your washing machine has flooded the laundry for the second time this year, and you’re standing there with a towel wondering whether this is the moment you stop fixing it.

The question is always the same, whatever the object: is this worth repairing?

Almost everything written about that question answers a different one. It tells you what a repair costs. It tells you how to do it yourself. It tells you which new model to buy. Very little of it tells you how to decide — how to take a repair quote, a machine’s age, and a fault, and turn those into a straight answer.

This page is that answer. It sets out the framework we use at RepairOrBin — five rules, applied in order — and it does so at a depth you won’t find elsewhere, because the framework is the thing that matters, not the individual appliance.

Everything below applies to any repairable object: a dishwasher, a television, a laptop, a lawn mower, a hot water system, a solar inverter. The numbers change. The logic doesn’t.

Why this framework exists
Repair-or-replace decisions are usually made on intuition. You get a quote, it feels like a lot, you compare it to what you remember a new one costing, and you decide. Sometimes that works. Often it doesn’t, and the failure goes in one of two directions.

Repairing something that should have been replaced. The quote was under half the replacement price, so it passed the test — but the machine was twelve years old, and the repair bought six months. Then the next fault arrived, and the next. You spent $600 over two years on a machine worth $200, and you replaced it anyway.

Replacing something that should have been repaired. The fault looked serious. The machine is old. The quote was $300, which felt like a lot for an appliance you’d had for years. You replaced it for $900 — when the actual fault was a $30 part and forty minutes of labour. This happens constantly, and it’s the more expensive mistake of the two.

Both failures share a cause: the decision was made without the two numbers that actually decide it. How much of the machine’s expected life has been used, and what the repair costs as a proportion of a comparable replacement.

Rules beat intuition because they force you to gather those two numbers before deciding. They also make the decision defensible — to a partner, to yourself, or to a repairer who’s quoting you $500 for something you’re not sure you need.

One thing this framework is not. It isn’t a way to justify replacing things. The rules reject repairs less often than intuition does, not more. A machine that fails the cost test frequently passes the lifespan test, and a cheap repair on an old machine often wins on the numbers even when it feels like throwing money away.

The two numbers, and a short reference table
Before the rules, you need a rough idea of both inputs. The full databases — lifespan by brand and model, repair cost by fault across every category — are a later phase of this page. For now, category-level figures are enough to apply the framework.

Average lifespan by category:

Category Typical lifespan
Dishwasher 9–13 years
Refrigerator 10–18 years
Washing machine 10–13 years
Dryer 10–15 years
Oven / range 13–18 years
Microwave 7–10 years
Television 7–12 years (LCD); 5–10 (OLED)
Laptop 4–7 years
Desktop PC 6–10 years
Smartphone 3–5 years
Air conditioner 12–17 years
Water heater 8–12 years (tank); 20+ (tankless)
Cordless drill 5–10 years
Lawn mower 8–12 years
Typical installed repair cost, against typical replacement cost:

Category Repair range Replacement range The repair that usually decides it
Dishwasher $70–$600 $400–$1,500 Control board
Refrigerator $100–$700 $700–$2,500 Compressor / sealed system
Washing machine $75–$500 $500–$1,500 Drum bearing
Dryer $75–$400 $400–$1,200 Motor
Oven / range $100–$600 $500–$2,000 Control board
Microwave $50–$300 $60–$500 Magnetron — and it’s usually a no
Television $100–$400 $300–$2,000 Panel — and panel is terminal
Laptop $80–$400 $400–$2,000 Motherboard
Air conditioner $150–$1,200 $1,500–$5,000 Compressor
Water heater $150–$600 $800–$2,500 Tank leak — terminal
The pattern worth noticing now, before the rules: in almost every category, there’s exactly one repair that costs more than half a new unit, and everything else is cheap. On a dishwasher it’s the control board. On a fridge it’s the sealed system. On a TV it’s the panel. On a laptop it’s the motherboard.

Once you know which fault you’re facing, the decision is usually already made. The rules exist to confirm it — and to catch the cases where the obvious answer is wrong.

The framework at a glance
Five rules, applied in order. The first four are the original RepairOrBin framework; the fifth was added because it kept deciding cases the other four missed.

Rule 1: The 50% Rule. If the repair costs more than about half a comparable replacement, replacement usually wins.

Rule 2: The Lifespan Rule. If the machine has used most of its expected life, replacement offers better value even when the repair passes the cost test.

Rule 3: The Reliability Rule. Multiple breakdowns in a short period mean replacement, regardless of individual repair costs.

Rule 4: The Safety Rule. Anything creating a safety hazard moves strongly toward replacement, whatever the cost. This overrides the other three.

Rule 5: The Obsolescence Rule. If the part, the software, or the standard no longer exists, the decision is made for you — and it’s usually replacement.

The order matters. Safety first, because it overrides everything. Obsolescence second, because if a part doesn’t exist there’s no cost to compare. Then cost, then age, then reliability — cheapest and most decisive first.

The rest of this page takes each rule in turn and explains how to apply it precisely, where it comes from, and where it fails.

Rule 1: The 50% Rule
If the repair costs more than about half the price of a comparable replacement, replacement usually wins.

This is the best-known rule of thumb in appliance repair, and it’s the one most people have heard of — but it’s also the one most often applied incorrectly. Three things need to be right for it to work: the comparison, the number, and the definition of “comparable.”

Where the rule comes from
The 50% threshold isn’t arbitrary. It’s an expression of a simple idea: a repair is only worth doing if it buys you more value than it costs. If a $300 repair extends a machine’s life by three years, and a replacement would cost $900 and last twelve, the repair costs you $100 per year of service while the replacement costs $75. The repair is worse value per year even though it’s cheaper in absolute terms.

At 50%, the repair is roughly break-even against replacement on a per-year basis — assuming the repair gives you a reasonable remaining life. That’s why the threshold sits where it does: it’s the point at which a repair stops being obviously better value than replacement.

Above 50%, you’re paying more per year of service than you would by replacing. Below it, you’re paying less. The rule is a proxy for that calculation, and it’s a good one.

Why 50% and not 30% or 70%
Lower thresholds reject too many repairs. A 30% rule would tell you to bin a machine whenever a repair costs a third of replacement — which would reject the door seal, the drain pump, the thermal fuse, and most of the cheap fixes that keep appliances alive for years. That’s not conservative, it’s wasteful.

Higher thresholds accept too many. A 70% rule would have you spending $600 repairing a machine you could replace for $900 — which fails on both per-year value and on the risk that the next fault is three months away.

50% is the point where the repair stops being obviously better value. It’s conservative by design, because it leaves room for the thing the calculation can’t predict: whether the machine fails again.

“Comparable” is the word that does all the work
The most common mistake in applying the 50% rule is comparing a repair to the cheapest replacement you can find. That’s not the comparison.

You’re comparing the repair to a machine with similar capacity, similar features, and similar quality. If your dishwasher has three racks, a cutlery tray, and a 15-place setting, the comparison isn’t a $400 budget machine with one rack. It’s the $900 machine that matches what you have.

This matters most at the premium end. A $400 control board repair on a $1,400 Miele is 29% — a repair. The same $400 against a $450 budget replacement is 89% — a bin. The number is identical; the comparison changes the answer entirely.

The practical test: if you were buying a replacement today, would you buy something cheaper than what you have, the same, or better? Price that. Most people replace like for like or slightly better, and that’s the figure the rule should use.

The call-out fee problem
This is where most repair-or-replace advice fails, and it’s the reason this framework exists in the shape it does.

A repairer charges to turn up and diagnose. In the US that’s typically $65–$165, and on many jobs it’s charged before a single part or hour of labour. That fee sits on top of every repair, and it changes the arithmetic on every small fault.

A $60 part becomes a $250 decision. A $150 repair becomes a $300 one. And the 50% test that the part would have passed comfortably now sits much closer to the line.

The rule to apply: use the quote, not the part price. Not the price of the part you found online, not the cost of the DIY repair if you’re not doing it yourself — the number on the invoice from the person who’d actually do the work.

This feels obvious, and people get it wrong constantly. The appeal of the $60 part price is that it makes the decision easier — a $60 fix against a $900 machine is obviously a repair, so you don’t have to think about it. But unless you’re fitting that part yourself, it isn’t the number that matters.

Worked examples across categories
Dishwasher, control board, eight years old. Quote $450. Comparable replacement $950. 47% — passes the cost test narrowly. But the machine is at 80% of expected lifespan. The lifespan rule rejects it. Bin.

Dishwasher, drain pump, seven years old. Quote $280. Comparable replacement $950. 29%. Machine at 70% of lifespan. Both rules pass. Repair.

Television, panel failure, four years old. Quote for panel replacement: $700. Comparable new TV: $800. 88%. Bin — and this is the case where the fault type alone decides it, because a panel always costs nearly as much as a new set.

Television, main board, four years old. Quote $180. Comparable new TV: $800. 23%. Repair.

Laptop, motherboard, five years old. Quote $400. Comparable new laptop: $900. 44%. Passes the cost test. But laptops have a 4–7 year lifespan, so the machine is at 70–100%. The lifespan rule rejects it. Bin.

Laptop, battery, four years old. Quote $120. Comparable new laptop: $900. 13%. Machine at 60–80% of lifespan. Repair — a battery is the cheapest way to extend a laptop’s useful life, and it’s almost always worth doing on anything under five years old.

Refrigerator, compressor, nine years old. Quote $700. Comparable replacement: $1,100. 64%. Fails the cost test. Bin — and the lifespan rule would have rejected it too.

Where the rule fails
The 50% rule is a heuristic, and there are three situations where it gives the wrong answer.

When the replacement has hidden costs. The quote for a repair is all-in. The price of a replacement usually isn’t — delivery, installation, disposal of the old unit, and any plumbing or electrical work can add 15–30% to the sticker price. If you’re comparing a $400 repair quote against a $700 sticker price, you’re not comparing like with like. Add the installation costs to the replacement side before applying the rule.

When the repair has a long warranty. A repair with a two-year parts-and-labour warranty is worth more than the same repair with no warranty, because it removes the risk of the next fault. This is one of the few genuine arguments for paying slightly more than the 50% rule would suggest — but only on the specific component, and only if the warranty is real.

When the machine is worth something on its own. A vintage appliance, a machine with sentimental value, or a unit that’s part of a fitted kitchen where a like-for-like replacement would require cabinetry work — these break the rule’s assumption that replacement is a simple swap. If replacing the machine means replacing the kitchen, the comparison isn’t the machine.

The DIY question
If you can do the repair yourself, the calculation changes substantially, because the part cost is often 20–40% of the installed cost. A $450 control board replacement becomes a $180 part. A $280 drain pump becomes an $80 part.

The rule doesn’t change — you still apply the 50% test — but you apply it to the number that’s actually yours to spend. Re-run the calculation against the part price, and most DIY repairs pass comfortably.

What DIY doesn’t change is the lifespan rule. A cheap repair on a machine at the end of its life is still a bet, even if you’re the one doing the work. The saving is in money, not in time.

Rule 2: The Lifespan Rule
If the product has already used most of its expected life, replacement offers better value even when the repair passes the cost test.

This is the rule that catches the case the 50% test misses. A repair that costs 20% of replacement passes the cost test comfortably — but if the machine is twelve years old and expected to last twelve, that 20% buys you a machine with no margin left.

The lifespan rule asks a different question: how much life is left, and is the repair buying you enough of it?

How lifespan figures are produced
Every lifespan figure on this page is an average of some kind, and it’s worth knowing what kind.

Consumer Reports publishes expected lifespans based on member surveys — how long appliances actually lasted in real households before being replaced or failing. These figures are useful because they reflect real usage rather than manufacturer claims.

The Association of Home Appliance Manufacturers (AHAM) publishes category averages based on manufacturer data and industry surveys. Their figures tend to run slightly higher than Consumer Reports.

Manufacturers publish their own expected service life, which is usually optimistic. Bosch says 6–16 years for their dishwashers, which is a range so wide it’s almost useless as a planning input.

Repair industry data — from parts suppliers, service networks, and repair chains — gives the most granular figures, because it tracks what actually fails and when.

What this means in practice: the figure you should use is a range, not a point, and the range is wide. A dishwasher doesn’t last ten years — dishwashers last somewhere between eight and fourteen, depending on the brand, the water, the usage, and the luck.

Ranges, not points
Every lifespan figure in this framework is a range, and the range matters more than the midpoint.

Category Conservative Typical Optimistic
Dishwasher 8 years 10 years 14 years
Refrigerator 10 years 14 years 18 years
Washing machine 8 years 11 years 14 years
Dryer 10 years 12 years 15 years
Oven / range 13 years 15 years 18 years
Television 7 years 9 years 12 years
Laptop 4 years 5 years 7 years
Smartphone 3 years 4 years 5 years
How to use the range: when deciding, use the typical figure as your baseline but ask which end of the range your machine is likely to be at. A well-maintained Bosch dishwasher with soft water is at the optimistic end. A budget machine with hard water, run twice a day, is at the conservative end.

The three thresholds
The framework uses three cut-off points, not one.

Below 50% of expected life: repair freely. The machine has more life ahead of it than behind. A repair is almost always the right answer if it passes the cost test, because you’re investing in something with plenty of service left.

Between 50% and 75%: repair carefully. The machine has more life behind it than ahead. A repair is still usually right, but the cost test needs to pass comfortably rather than narrowly, and the repair should be for a fault that’s likely to be a one-off rather than part of a pattern.

Above 75%: the repair needs a reason. Not a reason to replace — a reason to repair. The default shifts. You need the repair to be cheap relative to replacement, or the machine to be unusually well maintained, or the replacement to be unusually difficult or expensive. “It still works” isn’t a reason.

Past 100%: you’re in bonus time. The machine has exceeded its expected life. Every repair is a bet against the next failure, and the 50% test no longer captures the risk because the risk is no longer just financial. A repair can still be right — a $15 part on a twelve-year-old machine that’s otherwise sound is a reasonable bet — but it has to be cheap enough to be a bet rather than an investment.

What changes lifespan
The same appliance can last eight years or fourteen depending on five things, and they’re worth knowing because they tell you which end of the range you’re at.

Usage frequency. A dishwasher run once a day has roughly half the mechanical cycles of one run twice a day. A washing machine used by a family of five has a very different life from one used by a couple.

Water hardness. Scale is the single biggest killer of heating elements, water heaters, washing machines, and dishwashers. In hard water areas, expect the conservative end of the range unless you have a softener.

Maintenance. A machine with a clean filter, a clear vent, and a descaled element lasts longer than one that’s never been touched. This is the one factor you control, and it’s worth years.

Installation quality. A washing machine on an uneven floor, a fridge with no clearance for airflow, a dryer with a long restrictive vent — all shorten life. Poor installation is invisible until it isn’t.

Brand and design. Premium brands are built to last longer, and it’s not only marketing. Miele’s 20-year design life on dishwashers is a genuine engineering claim, reflected in the failure rates. The budget end of the market is designed to a price, and lifespan is one of the things that price buys.

Brand variance
The category ranges hide a bigger variance between brands than within them. On dishwashers, the range across all brands is 8–14 years — but a Miele at 8 years is a different proposition from a Frigidaire at 8 years.

Brand tier Lifespan adjustment Examples
Premium built-to-last +30–50% on the category average Miele, Sub-Zero, premium Bosch
Mid-premium +10–20% Bosch, KitchenAid, higher-end Whirlpool
Mid-range Category average GE, Whirlpool, Maytag, LG, Samsung
Budget -10–20% Frigidaire, Amana, budget store brands
This is where the framework and the category tables meet. A twelve-year-old Frigidaire dishwasher is 100–150% of expected life. A twelve-year-old Miele dishwasher is 60–80%. The same age, the same category, a completely different decision.

Worked examples
**Six-year-old Bosch dishwasher, $300 repair.** Typical lifespan 10–15 years for Bosch. Life used: 40–60%. Below the 75% threshold. **Repair**, and the cost test passes at 32% of a $950 replacement.

Twelve-year-old Frigidaire dishwasher, $150 repair. Typical lifespan 8–12 years for Frigidaire. Life used: 100–150%. Bonus time. The repair is cheap, so it’s a reasonable bet — but if a second fault arrives within a year, the answer flips to replacement.

Nine-year-old Miele dishwasher, $400 repair. Typical lifespan 15–20 years for Miele. Life used: 45–60%. Repair, comfortably, and the premium brand makes this an easy call even at 40% of replacement cost.

**Four-year-old laptop, $350 motherboard repair.** Typical lifespan 4–7 years. Life used: 57–100%. The upper end of the range is past the 75% threshold. **Lean bin**, unless the laptop is a high-spec machine that would cost $1,800 to replace.

The honest limitation
Lifespan figures are averages of other people’s machines. Yours is one machine, and it will either fail early or outlast the average — you can’t know which in advance.

What the rule does is price the risk. A repair on a machine at 40% of expected life is a bet with good odds. The same repair at 120% is a bet with poor odds. The rule doesn’t tell you which way the bet will land; it tells you whether the odds justify taking it.

Rule 3: The Reliability Rule
Multiple breakdowns in a short period mean replacement, regardless of the individual repair cost.

One repair is a repair. Two repairs in two years is a pattern. Three is a machine telling you something.

This rule exists because the 50% test is applied to a single repair, and real appliances fail in clusters. A machine that needs a pump, then a seal, then a board is doing exactly what old machines do — and each individual repair will pass the 50% test while the sum of them doesn’t.

The threshold
The practical threshold the framework uses: two significant repairs in three years, or three in five.

That’s not a hard number, and it’s not sourced from a study — it’s a working threshold that captures the pattern. The logic is that a machine which fails twice in three years is on a declining trajectory, and the next failure is more likely to arrive sooner than the last one did.

What counts as significant. A repair that required a part and either a technician or a substantial DIY effort. A blocked filter that you cleared yourself doesn’t count. A failed drain pump does. The threshold is about component failures, not maintenance.

The logic of clusters. Failures on an appliance aren’t random. They cluster around age, and they cluster around shared causes. A failed door seal lets water into the drum bearing. A blocked vent overheats the thermal fuse and the heating element. A failing control board causes three apparently separate faults. When you see two failures close together, the third cause is often already in progress.

Why clusters matter more than individual costs
Here’s the case the 50% rule misses entirely.

A washing machine at seven years needs a drain pump. Quote $250. Against a $900 replacement, that’s 28% — a clear repair by the cost test.

Eight months later, the same machine needs a door seal. Quote $200. Against $900, that’s 22% — another clear repair by the cost test.

Fourteen months after that, the drum bearing fails. Quote $450. Against $900, that’s 50% — borderline, and by then you’ve spent $450 on the first two repairs.

Total spent on repairs: $900. Total saved by not replacing: nothing. The individual decisions were each defensible. The cumulative outcome was the same as buying a new machine, but without the new machine.

The reliability rule catches this. After the second repair in three years, it says stop — and that’s the point at which replacement would have saved you the third repair and the eighteen months of uncertainty.

Worked examples
Dryer, seven years old, first fault. Thermal fuse blown by a blocked vent. $150 installed. First repair in the machine’s life. Repair, and clear the vent to prevent the recurrence.

Dryer, same machine, nine months later. Heating element failed. $250 installed. That’s two significant repairs in under a year. The reliability rule says: start pricing replacements. The second repair is still justified — the machine is nine years old and otherwise sound — but the next fault should tip it.

Refrigerator, eight years old, third repair in four years. Evaporator fan, then defrost heater, then thermistor. Individually $250, $200, $180. Cumulative $630 against a $1,100 replacement. The reliability rule has been triggered for a while. Bin.

Television, three years old, first fault. Main board. $180. One repair. Repair, comfortably.

When the rule doesn’t apply
On a machine with an unusually long design life. A Miele dishwasher with two repairs in four years at year eight is still a machine with a decade of life ahead. The reliability rule assumes typical lifespan — premium machines get more latitude.

When the faults are unrelated and cheap. A door seal and a drain filter are both wear items, not signs of decline. If both repairs were under $150 and neither required a technician, the pattern is less meaningful.

When a known design fault has been fixed. Some models have specific recurring faults that are fixed by the first repair. If the repair addressed a design weakness and the machine is otherwise sound, one repair isn’t a pattern.

Rule 4: The Safety Rule
Anything that creates a safety hazard moves strongly toward replacement, whatever the cost.

This rule overrides the other four. It’s not a percentage, and it doesn’t get weighed against the economics. A machine with a safety fault stops being a repair-or-replace decision and becomes a decision about whether to keep using it at all.

The stop-immediately list
These are the cases where the answer is not “repair” or “replace” — it’s “stop using it.”

Burning smells during operation. Not a hot smell — a burning smell. Plastic, wiring, or an electrical burning smell. Stop the machine, unplug it, and have it assessed.

Scorch marks or discoloured wiring. Any visible heat damage on a power cord, plug, or internal wiring. This is a fire risk and it’s not a DIY repair.

Gas smells from a gas appliance. Turn off the supply, ventilate, and call a licensed gas fitter. Do not run the appliance, and do not attempt to diagnose it yourself.

Water on electrical components. A leak that’s reached the wiring, the control board, or the motor. The machine needs to be fully dry and inspected before it runs again.

A swelling, hot, or damaged lithium battery. Laptop, phone, tablet, cordless tool, vacuum. A swollen battery is a fire risk and should not be charged, used, or opened. Dispose of it at a battery recycling point.

A tripping circuit breaker that resets and trips again. A short circuit. Stop using the appliance and have it assessed.

Physical damage to a gas appliance or a water heater tank. A dented gas line, a leaking water heater tank, a cracked heat exchanger. These aren’t repairs.

Recalls and known faults
A recalled appliance is a safety decision, not a repair decision. Check your model against the manufacturer’s recall notices and any national recall register before spending money on a repair — if the machine is subject to a recall, the manufacturer may repair or replace it for free.

Some known design faults are safety-relevant even without a formal recall: certain fridge models with compressor issues, certain washing machines with door lock faults, certain dryers with lint trap designs that increase fire risk. If a fault on your machine matches a documented safety issue, treat it as a safety decision.

When safety and economics disagree
The most difficult cases are the ones where the safety issue is real but the machine is valuable and repairable.

A gas oven with a faulty igniter that’s leaking a small amount of gas before it lights — the repair is $150 and the machine is worth $1,200. The economics say repair. The safety rule says: stop using it until it’s repaired. Not “replace it” — repair it, but immediately, and don’t use it in the meantime.

The rule isn’t “always replace when there’s a safety issue.” It’s “the safety issue takes priority over the cost calculation.” Some safety issues are repaired. Some end the machine’s life. None of them get weighed against a percentage.

The categories of hazard
Electrical. Burning smells, scorch marks, tripping breakers, damaged cords, water ingress. The largest category, and the one most often ignored because the machine still works.

Gas. Leaks, incomplete combustion, faulty igniters, damaged lines. Never a DIY diagnosis.

Fire. Dryer lint accumulation, blocked vents, overheated heating elements, lithium batteries. The dryer vent is the single most common appliance fire cause, and it’s free to prevent.

Water damage. A leak that reaches electrical components or causes structural damage. The machine may be repairable, but the damage it’s caused may not be.

Physical injury. Sharp edges, heavy components, spring-loaded parts, glass. Not usually a replace-decision, but a reason to stop and get help rather than continue alone.

Chemical and refrigerant. Refrigerant leaks, chemical spills in washing machines, mould in seals and ducts. Some are health issues rather than safety ones, but they move the decision the same way.

Rule 5: The Obsolescence Rule
If the part, the software, or the standard no longer exists, the decision is made for you — and it’s usually replacement.

This is the newest rule in the framework, and the one that’s become more important every year. It covers the cases where the economics are irrelevant because the repair isn’t available.

Parts availability
The first question on any repair isn’t “how much” — it’s “does the part exist.”

Manufacturers are required to supply spare parts for a period after a product is discontinued, but the period varies by region and by product. In the EU it’s typically 7–10 years for major appliances. In the US there’s no equivalent requirement, and parts availability depends entirely on the manufacturer.

The practical consequences:

Discontinued models. A machine that’s out of production may have no parts supply beyond whatever is sitting in a warehouse. A $50 part becomes unavailable, and a $500 repair becomes impossible.

Integrated assemblies. A control board that used to be a $60 part is now integrated into a $300 assembly, because the manufacturer consolidated the design. The repair is technically possible and economically unviable.

Proprietary fasteners and sealed units. A machine designed so the consumer can’t open it is a machine designed to be replaced. This isn’t a fault — it’s a design decision, and it decides the repair-or-replace question before you start.

What to check before pricing a repair: search for the part by model number. If it’s available from multiple suppliers at a reasonable price, the repair is viable. If it’s only available from one source at an inflated price, or not available at all, the economics don’t matter.

Software and firmware support
An increasing number of appliances and devices have software that matters to their function — smart TVs, connected appliances, laptops, phones, tablets.

When software support ends, the device is functionally obsolete even if the hardware works. A smart TV that no longer receives app updates is a TV that can’t run Netflix. A phone that no longer receives security updates is a phone that shouldn’t be used for banking. A laptop that can’t run a supported operating system is a laptop that can’t do the work.

The lifespan figures for electronics already reflect this — a smartphone’s 3–5 year lifespan is driven more by OS support than by battery or screen failure.

The obsolescence question: is the software still supported, and will it be for the period the repair would buy? If a $200 laptop repair buys two more years, but the OS goes out of support in eighteen months, the repair is buying eighteen months, not two years.

Standards changes
Some repairs are made uneconomic by changes to standards rather than by parts.

Refrigerants. The phase-out of R22 and the transition to R32, R290, and R1234yf means older air conditioners and fridges use refrigerants that are increasingly expensive and restricted. A compressor repair on an R22 system may be possible but not economic.

Energy efficiency standards. Older appliances may be so inefficient that the running cost exceeds the repair saving. A 15-year-old fridge can use twice the electricity of a current model — which changes the per-year value calculation in the 50% rule.

Electrical standards. Older homes and older appliances sometimes sit on different standards — two-wire outlets, ungrounded circuits, older plug types. Adapting an old appliance to a current standard can cost more than the appliance.

Connectivity and compatibility. An older smart device that only supports Wi-Fi 4, or an older TV that only supports HDCP 1.4, may not work with current equipment. The repair might fix the fault while leaving the machine incompatible with everything around it.

Repairable by design
The inverse of obsolescence is worth knowing, because it’s the thing to look for when buying.

A machine is repairable by design when:

Parts are available separately rather than only as assemblies

The service manual is published

The fasteners are standard rather than proprietary

The components are accessible without dismantling the whole machine

The manufacturer supports a repair network

This is the right-to-repair question, and it has a practical consequence: a repairable machine costs less over its life than a non-repairable one, even if it costs more up front. A $1,200 machine that’s repairable for fifteen years is cheaper than a $600 machine that’s binned at eight.

When the rule applies
A discontinued appliance with an unavailable part → replacement, regardless of cost.

A device with unsupported software → lean replacement, because the hardware repair doesn’t restore function.

An appliance on a phased-out refrigerant → replacement, because the repair cost will rise and the parts will become scarcer.

A machine with an integrated assembly where a component has failed → price the assembly. If it exceeds the 50% line, the obsolescence rule and the cost rule agree.

Applying the framework
The order of operations
The rules are applied in a fixed order, and the order matters more than any individual rule.

1. Is it safe? If not, stop using it. Then repair or replace based on the other rules — but not until it’s safe.

2. Does the part exist, and is the software supported? If not, the decision is made. Replace.

3. Is the repair under 50% of a comparable replacement, using the full quote? If not, replace.

4. How much of the expected lifespan has been used? Under 50%, repair. 50–75%, repair if the cost test passes comfortably. Above 75%, the repair needs a reason. Above 100%, it’s a bet.

5. Has it needed multiple repairs recently? Two in three years, or three in five, tips toward replacement.

If all five pass, repair. If any one fails decisively, replace.

The flowchart
Start.

↓ Burning smell, gas, scorch marks, swelling battery, water on electrics?
→ Yes: Stop. Do not use. Assess or replace.

↓ Is the part available, and is the software supported?
→ No: Replace.

↓ Is the repair quote under half the cost of a comparable replacement, including installation of the new unit?
→ No: Replace.
→ Yes: continue.

↓ Has the machine used more than 75% of its expected life?
→ No: Repair.
→ Yes: continue.

↓ Is the repair cheap enough to be a bet rather than an investment — under 25% of replacement?
→ Yes: Repair, and accept the risk.
→ No: Replace.

↓ Has it needed two significant repairs in three years?
→ Yes: Replace.
→ No: Repair.

A worked example, end to end
A ten-year-old Whirlpool dishwasher, not draining. Standing water in the bottom. The filter is clear, the drain hose is clear, and the pump impeller turns freely by hand but the pump doesn’t run during a cycle.

Safety: no burning smell, no scorch marks, no water on electrical components. The water is contained in the tub. Safe to proceed.

Obsolescence: the drain pump is available from multiple suppliers at $40–$90. The model is still supported. Not obsolete.

Cost: installed quote $280. Comparable replacement: a similar-capacity Whirlpool at $750, plus $100 delivery and installation, plus $40 disposal — call it $890. $280 is 31%. Passes.

Lifespan: Whirlpool dishwashers typically last 9–12 years. At ten years, the machine has used 83–111% of expected life. Past the 75% threshold. The repair needs a reason.

Is the reason there? The repair is $280 against an $890 replacement — 31%, which is above the 25% “bet” threshold. The machine is at the end of its expected life. There’s no particular reason to expect another five years from it.

Reliability: this is the first repair in five years. No pattern.

The decision: the cost test passes, but the lifespan test doesn’t, and the repair is too expensive to be a cheap bet on an old machine. Replace — though a repair would be defensible if the machine were otherwise in excellent condition and the budget didn’t allow a replacement right now.

What would change the answer: if the machine were seven years old rather than ten, the repair would win comfortably. If the quote were $150 rather than $280, the cheap-bet threshold would be met and the repair would be reasonable. If it were the second repair in two years, the reliability rule would confirm the replacement.

What’s coming next
This page is being built in phases, and this is Phase 1 — the framework itself, at depth. The remaining phases will add:

The databases — appliance lifespan by brand and repair cost by fault, across every category. These are the tables that make the framework actionable at a specific machine rather than a category average.

Phase 3: The category sections. Appliances, electronics, tools, home systems — each one at the depth the Appliances section will be, with the decision repair named for every product, brand notes, and worked examples.

Phase 4: The calculator. An interactive tool that takes your repair quote, your machine’s age, and its category, and returns a recommendation. The tables come first; the tool is the front end.

In the meantime, the framework above works. The category tables in this phase give you enough to apply all five rules to any common appliance, and the rules themselves don’t change when the data gets more granular.

A note on the numbers
Every figure on this page is in US dollars, and every figure is a range rather than a point.

The ranges are real. Repair costs vary by region, by brand, by whether you’re in a city or a rural area, and by whether the technician is independent or part of a national chain. A $200 repair in one market is a $350 repair in another for the same fault.

The sources are mixed. Lifespan figures draw on Consumer Reports, AHAM, and manufacturer data. Repair costs draw on HomeAdvisor, Fixr, and published service pricing. Where sources disagree, the page gives the range rather than picking one.

The framework is the durable part. The numbers will date. The five rules won’t — they’re a way of thinking about the decision, and they’ll apply to whatever the repair costs are in ten years’ time.

If you’re outside the US, the percentages still work. Convert the numbers to your currency and apply the same rules — the ratio between repair cost and replacement cost is what matters, not the absolute figure.

Find the fault first

Not sure yet? The Repair or Bin Framework Calculator applies all five rules to your specific quote in under a minute.

The framework above assumes you know what’s wrong. If you don’t yet, the fault guides are the starting point — every one of them diagnoses first and gives a repair-or-bin verdict at the end.

Kitchen: Dishwasher hub · Fridge leaking · Not cooling · Oven not heating · Microwave not heating · Air fryer not working

Laundry: Dryer pillar · Washer not spinning · Not draining · Leaking

Electronics: LG TV hub · Laptop won’t boot · LG TV fault diagnosis

Outdoor and home: Ceiling fan wobbles · Gas grill igniter · Dyson pulsing

Browse everything: All repair guides →