Repair or replace? The honest maths on fixing computers
The industry default is "just replace it", and it exists because replacement is easy to sell, not because it's usually right. A big share of the machines that end up in a skip needed one cheap, boring fix: a drive, a battery, a power jack. The maths deserves better than a reflex, and so does the e-waste pile. Here's the framework we actually use: the repairs that are nearly always worth it, the ones that rarely are, and the full formula everyone gets wrong because they forget the cost of moving to the new machine.
Why the default answer is always "replace"
Walk into most shops with a slow laptop and you'll walk out with a quote for a new one. Not because the old machine was assessed and failed, but because selling a replacement is quick, clean and better margin than an honest diagnosis. Repair takes bench time and skill, and it ends in a smaller invoice. The incentives all point one way, so the advice points the same way. That's not a conspiracy, it's just how the industry is set up, and it's why "is it worth repairing?" so rarely gets a straight answer.
Repairs that are almost always worth it
Some jobs are such reliably good value we barely need to think about them:
- SSD and RAM on a structurally sound machine. If the computer still runs a spinning hard drive, swapping it for an SSD and topping up the memory is the single best-value fix in computing. The machine feels new, because for everyday work it effectively is. Most "it's gotten so slow" complaints end right here.
- Batteries. A battery is a consumable, like tyres on a car. A laptop that dies by lunchtime isn't worn out. It just needs the consumable replaced.
- Screens on mid-tier and better machines. A cracked screen on a decent business laptop is cosmetic damage on a healthy computer. The panel swaps out and the rest of the machine never knew it happened.
- Power jacks and charging faults. Wiggle-the-cable syndrome. Usually a worn connector, a small and well-understood repair on an otherwise fine machine.
The common thread: the fault lives in one replaceable part and the bones of the machine are good.
Repairs that usually aren't
The honest list cuts the other way too:
- Motherboard failures on old consumer laptops. The board is most of the machine's value, boards for ageing consumer models are hard to source, and the labour costs the same whether the laptop was cheap or premium. On a budget machine past its prime, this maths almost never works.
- Liquid damage. It's roulette. A cleaned board might run for years, or corrosion can keep creeping and kill it a month after you paid for the repair. We'll say that plainly up front: recover the data first, then decide with clear eyes.
- Anything where parts cost approaches replacement cost. Once the quote for parts alone sits near the price of an equivalent machine, stop. You'd be paying new-computer money for an old computer.
The real formula, including the bit everyone forgets
The comparison most people run is the repair quote against the sticker price of a new machine. That's half the maths. The full version:
Repair cost plus remaining useful life, versus replacement cost plus migration time.
Migration time is the forgotten half. A new computer isn't working the moment you unbox it. Data has to move. Software has to be reinstalled and reactivated: licences, printers, email profiles, that one accounting package whose licence key nobody can find. For a business machine, add the hours the person using it loses to "where did that setting go". A repaired machine skips all of it. Same machine, same setup, back on the desk.
Remaining life is real too, and it cuts against repair. A fix that buys six months on a machine already struggling with modern work is bad maths even when the quote looks small. The question is never "which is cheaper today". It's "which is cheaper per year of useful machine".
The age rule of thumb, honestly
The trade rule of thumb puts the line around five years for a business machine, and as rules of thumb go it's not bad. But it needs honesty in both directions. A well-built business-grade laptop is serviceable all the way through, so with an SSD and a fresh battery it can earn its keep well past the line. A bottom-shelf consumer laptop can fail the test at three, because it was built to a price and no repair fixes that. And the hard cutoff is software, not hardware: once a machine can no longer run a supported operating system with current security updates, repair is off the table for anything that touches business data.
The business fleet angle
For a business the calculation changes shape. Repair is what keeps a standard fleet standard: same models, same setup, shared spares, and one dead laptop doesn't force an odd new machine into the mix. If your fleet is consistent, repairing to preserve that consistency is usually the smart move, and a machine you retire can become the parts shelf for its siblings.
But there is a genuine tipping point where a planned refresh beats endless patching: when machines start falling off supported operating systems together, when warranty is gone across the board, or when staff are visibly waiting on their tools. A refresh done deliberately, in one planned hit with proper migration, costs less in disruption than replacing the same machines one emergency at a time. The worst outcome is neither: a fleet patched forever by default because nobody ever ran the numbers.
The e-waste point
Australia's e-waste problem is real, and computers binned with a cheap fix still in them are an avoidable part of it. You don't need to be an environmentalist for that to bother you. A working machine in a skip is wasted money as much as wasted material, and most of a computer's environmental cost was spent making it, so every extra year of service counts.
Repair-first is the biggest lever, but not the only one. A machine that isn't worth fixing whole is often worth harvesting: its RAM, drive and screen can keep others of the same model alive. When it truly is dead, e-waste recycling exists exactly for this, and a computer never belongs in the household bin. One non-negotiable before any machine leaves your hands: wipe the drive properly or pull it out, because your files don't stop being your files just because the computer stopped booting. The data side is covered in our backup guide.
How we handle it
Simple, and it's what we'd want as customers ourselves. We assess the machine honestly, quote the repair and the replacement side by side, and tell you which one we'd pick if it were our own gear and our own money. Sometimes that's the repair. Sometimes it's genuinely the new machine, and when it is, we say so and handle the migration properly so the hidden cost stays small. If it lands on replace, our procurement service covers choosing and rolling out the right machine without the guesswork.
FAQ
Is my laptop worth repairing?
Usually yes, if the machine is structurally sound and the fault sits in one replaceable part: drive, memory, battery, screen or power jack. An SSD and RAM upgrade on an otherwise healthy laptop makes it feel new for a fraction of the cost of replacing it. It stops being worth it when the motherboard fails on an older consumer machine, or when the parts quote approaches the price of an equivalent replacement.
When should a business replace computers instead of repairing them?
When the platform is done rather than the part: machines falling off supported operating systems, warranty expired across the fleet, staff losing time waiting on their tools. At that point a planned refresh in one deliberate hit, with proper migration, beats replacing the same machines one emergency at a time. Until then, repair is what keeps a standard fleet standard.
What computer repairs are usually not worth paying for?
Motherboard replacements on older consumer laptops, because the board is most of the machine's value and the labour costs the same regardless. Liquid damage is roulette: it can run for years after a clean-up or die a month later from spreading corrosion. And any repair where the parts quote sits near the price of an equivalent machine. In all three cases, recover the data first, then decide.
How old is too old to repair a computer?
The rule of thumb says around five years for a business machine, but it needs honesty both ways. A well-built business-grade laptop can be worth fixing past that line, because everything in it is serviceable. A cheap consumer laptop can fail the test at three. The hard line is software: once it can't run a supported operating system with security updates, stop repairing it for business use.
What should I do with a computer that is genuinely dead?
Three steps. Wipe or remove the drive first, because your data is still on it whether or not the machine boots. Harvest anything useful: RAM, drives and screens from a dead machine can keep its siblings alive, which matters if you run several of the same model. Then send the rest to a proper e-waste recycler rather than the bin, since computers hold both recoverable materials and things that should never reach landfill.
Does repairing computers actually reduce e-waste?
Yes, and it's the biggest lever available. Most of the environmental cost of a computer is in manufacturing it, so every extra year of service from an existing machine is a year another one doesn't have to be built. Australia's e-waste problem is real, and machines binned with a cheap fix still in them are an avoidable part of it. Repair first, harvest second, recycle last.
The bottom line
Repair or replace is a calculation, not a reflex. Fix the machine when the fault is one good part on solid bones. Replace it when the platform itself is done, and do that deliberately rather than one emergency at a time. Run the full maths including migration time, and let the machine that still works keep working. Your budget and the e-waste pile both come out ahead.
Got a machine sitting on the fence? Bring us the symptoms and you'll get both quotes and a straight recommendation: the repair, the replacement, and which one we'd pick if it were ours. Tell us what it's doing and we'll take it from there.