The compressor quote comes back at $640. The machine is eleven years old and might fetch $800 from the right buyer. Everyone in the room knows the answer, and the answer is to replace it.
That is not the question. The question is what the next few years cost if you keep it, set against what they cost if you swap it. Two ledgers, and most operators fill in one line of one.
I have watched this go wrong both ways. A fourteen year old Crane snack machine got a $150 coin mechanism repair and ran four more years. A six year old unit got a $900 control board and was back on the bench five months later. So when to replace a vending machine is a question about windows and locations, not birthdays.
Age Is a Bad Place to Start
Industry ranges are wide enough to be useless on their own. A machine in a hot garage with a cracked door gasket and a condenser packed with lint is done at six or seven years. The same model, cleaned quarterly and kept level, runs past twenty. The spread between worst and best is roughly three to one, and most of it comes down to two things that cost almost nothing. Coil cleaning and door gaskets.
What age is good for is one narrow job. It tells you whether parts are still made. Once a model leaves a manufacturer’s support list, every repair becomes a search rather than an order, and the search is where the money goes. A $70 motor that ships in two days is a nuisance. The same motor hunted through three resellers over nine days is a lost location.
So treat age as a parts check, not a verdict. Then watch three signals, none of them the repair quote.
- How many service calls in the last twelve months, and were any of them the same fault twice
- How many days the machine was actually out of service
- Whether it can still take a card without an argument
A machine that goes down more than a day a month is costing you more in quiet ways than in invoices. Downtime is the only line that never arrives as a bill.
The 50 Percent Rule and Where It Breaks
The rule most operators use is simple. If a repair costs more than half of what the machine is worth right now, replace it. A ten year old machine with a market value near $800 should not get a $500 compressor overhaul when three other parts are the same age and the same distance from failing. A second version compares the repair to the price of a new machine with a 40 to 50 percent trigger.
Both are useful first passes and both fail in the same place. They compare one invoice to one purchase and ignore everything between.
The rule says nothing about what the machine earns. A tired machine pulling $900 a month at a location you fought for is a different asset than the same machine in a break room doing $180. Same quote, same market value, different answer. And it treats the old machine’s market value as fact when it is usually a guess. A number from a listing site is not an offer. If nobody has bid, treat that $800 as $500 and see whether the answer changes.
Six Numbers on One Ledger
This version holds up, and it turns when to replace a vending machine into a number rather than a feeling. Six numbers, all measured over the same window, all on the same side of the comparison. The window decides the outcome more than any of the six.
One. The quote in front of you
The only number most people use. Write it down and set it aside, because on its own it is the least informative line on the page.
Two. The repairs you have not had yet
A machine that failed once will fail again, and the second failure is not in the quote. Take your repair history, work out how often a major component goes, and price the expected number of them over the window. One significant failure in eighteen months means planning for one or two more across two years. That is not pessimism, it is the distribution repeating.
Compressors, control boards, and refrigeration work sit in a different bracket from a spiral motor. On modern machines with Android based controls, a replacement board alone runs $300 to $500 and often has to come from the factory.
Three. What the service call really costs
The invoice is not the cost. The median wage for maintenance and repair workers sat at $23.38 an hour in the most recent federal survey, before travel, diagnostics, and the second visit that happens when the first one guesses wrong. A call-out with an hour and a half of labor and a $70 part commonly lands near $290.
Count visits, not repairs. Routine maintenance runs $200 to $600 a machine per year, and an old machine sits at the top of that range while a new one sits near zero through its warranty.
Four. Parts and payments
Two risks bundled together because they both kill old machines.
Parts first. When a manufacturer stops supporting a model, availability dries up and prices climb. Regulators have started naming restricted diagnostic software and repair-hostile designs as barriers that raise lifecycle risk.
Payments second, and this one catches people who think their machine is mechanically fine. The major carriers finished shutting down their 3G networks during 2022, so a cabinet can be in perfect order while the modem inside it is a paperweight. Readers also carry security requirements that get revised, and physical fit is not the same as support. A machine that cannot take cards is losing sales whether or not it vends. In one large cashless dataset, card and mobile payments accounted for roughly 71 percent of vending transactions, so a cash-only machine is not selling to a smaller market, it is selling to a different one.
Retrofitting is possible and sometimes sensible. A reader and cashless kit runs $300 to $550 plus commissioning, and on a sound machine that can add five years. On a machine with a tired compressor it is money spent on the wrong end.
Five. The energy difference
Older refrigerated machines commonly draw two to three times the electricity of a current unit. On a machine running continuously the monthly gap lands around $8 to $20 depending on rates. Over two years that is $190 at the low end and close to $480 at the high end. It always favors replacement, which is why salespeople quote it and operators skip it. Include it, but do not let it decide anything alone.
Six. The revenue the new machine would earn
Almost nobody puts this in, and it usually flips the comparison.
A new machine at a good location does not earn what the old one earned. It takes cards without failing, it holds a planogram you have since learned is better, and with telemetry it gets restocked on data rather than a guess. A ten percent lift is conservative at a location with real traffic.
Put it in the ledger as margin, not gross. At $900 a month and a fifty percent margin, ten percent is $45 a month, or $1,080 across two years. That number does not exist on the repair side of the page, and it is why the arithmetic so often refuses to agree with the gut.
Which Window You Use Decides the Answer
The same machine, the same six numbers, two different windows.
An eleven year old snack machine in a plant with about 180 people. Nine hundred dollars a month gross, fifty percent margin, so $450 of contribution. A $640 compressor quote, a market value near $800, a replacement delivered and commissioned at $5,200.
Keep it, two years. Compressor $640. One expected further failure at $550 times a 60 percent chance, $330. Two routine visits at $180, $360. Energy penalty about $190. Lost margin from roughly twelve days down, about $180. Total near $1,700.
Replace it, two years. $5,200 for the machine, less $800 for the old one, less $190 of energy, less $1,080 of added margin. Total near $3,130.
Repair wins by about $1,400. The gut was wrong.
Now stretch the window to five years, which is what you are doing if you plan to keep the location.
Keep it, five years. The compressor, the same $330 of expected failures, five routine visits at $180 so $900, energy at $475, lost margin at $450. Total near $2,795. This column also hides something. At sixteen years old the machine still has to be replaced at the end of it, at whatever a machine costs then.
Replace it, five years. $5,200, less $800, less $475 of energy, less $2,700 of added margin, plus about $400 of repairs after the warranty ends. Total near $1,625.
Replacement wins by roughly $1,170, and the crossover sits in year four.
So the honest answer to when to replace a vending machine is that it depends on how long you intend to own the location. On two years repair almost always wins on cash. On five, replacement usually does. Anyone giving you one answer without asking about the window is selling something.
The Real Reason to Replace Early
The ledger rarely justifies replacing a machine that still works. What justifies it is the location.
A tired machine does not fail evenly. It fails in front of the person who decides whether you keep the spot. Three days dark in a quarter is survivable. Three days dark twice in a quarter becomes a conversation you did not schedule, and that conversation is where locations are lost.
Price that. A $900 a month location is $10,800 a year of gross, and losing it costs the machine’s whole remaining contribution plus the cost of finding another. Against that, the $1,400 gap in the two year ledger is not a gap. It is insurance at about $58 a month, and that is the argument to have with yourself when the numbers say repair. Not whether the machine is worth fixing, but whether the location is worth protecting from it.
The Third Option Nobody Prices
Between repair and replace there is a middle move that gets almost no attention. Move the tired machine to a lower value location and put the new one where the money is.
A break room doing $180 a month does not need a reader that never fails or a planogram tuned to foot traffic. It needs a cabinet that holds snacks and someone who shows up. The high traffic location needs the reliable machine, because that is where downtime costs money and where a manager is watching.
That single move changes the arithmetic. You buy one machine, not two, and the old one keeps earning somewhere it is not on trial. The only cost is a truck, a dolly, and an afternoon.
Two rules. The old machine has to be safe, meaning no structural rust, no exposed wiring, and a door that latches. And it still has to be serviceable, meaning parts exist. A machine you cannot fix is not a spare, it is a future apology.
A Replacement Queue, Not a Replacement Decision
Past a handful of machines, when to replace a vending machine stops being a decision about one unit and becomes a question of order.
Rank by what each location earns per month, then adjust for two things. How much downtime the machine has caused, and whether the location contact has ever mentioned it. A machine nobody has complained about can wait even if it is old. A machine mentioned twice cannot, even if it is younger.
Write the queue down with a year beside each one. A five year plan with one machine a year is a budget line you can argue with. A surprise failure is an emergency purchase at whatever is available this week.
One more thing. Planned replacement lets you sell the old machine while it still runs, and a working machine commands a real price while a dead one scrap weight. Pull the cash box and the reader before it leaves, and photograph the serial plate.
Where This Leaves You
We build vending machines, so treat our bias as bias. The honest position is that most machines get replaced a year or two before the arithmetic demands it, and the reason is almost never the machine.
If you are weighing a replacement and want to know what a new machine would do at your location, tell us the traffic and the product mix and we will work through the numbers with you.
Perguntas frequentes
How long does a vending machine last?
Between five and seven years in poor conditions, ten to fifteen with ordinary maintenance, and past twenty with careful upkeep. Condenser cleaning, door gasket condition, and a level floor decide most of it. Refrigerated units are more sensitive to all three.
Is it worth repairing a twelve year old vending machine?
It depends on parts availability and what the machine earns. If the failed part is modular and still in production, and the location produces steady revenue, a repair is often cheaper over two years. If the part has to be hunted through resellers, or the machine cannot take modern cards, the repair only buys time.
What is the 50 percent rule for vending machines?
If a repair costs more than half of what the machine is worth right now, replacement is usually better. A second version compares the repair to the price of a new machine with a 40 to 50 percent trigger. Both are first passes, and both ignore what the machine earns.
Can I add a card reader to an old machine instead of replacing it?
Sometimes, and it is often the best value on an older machine that is otherwise sound. A reader and cashless kit typically runs $300 to $550 plus commissioning, and it removes the coin mechanism from the failure list while opening the machine to buyers who no longer carry cash. Confirm the machine supports the standard interface before buying anything.
What should I do with a vending machine I am replacing?
Sell it privately if it runs, consign it with a dealer if you want the process handled, keep it as a spare for a lower traffic location, or scrap it if it does not run. Sell before it dies, not after. And do not donate it to a location and call the relationship handled, because it still has your name on the service sticker when it fails.
