Ten Grand, Two ATMs, and the Money That Never Leaves the Box

Here is the first thing nobody tells you about putting $10,000 into ATMs: a third of it or more never gets to work. It sits in a metal cassette in the back of a bar, waiting for someone to need forty bucks. It is not spent. It is not earning. It is just parked, and it is the price of admission.

I ran a vending machine business before I sold it, and I still get asked about the cousins of vending, ATMs especially. So I want to walk through what ten grand really buys in an ATM business, where the money goes, and the question I would ask myself before writing the check. I have not run an ATM route myself, so this is the math and the reasoning, not a war story.

Disclosure: this post contains an affiliate link. If you sign up through it, I may earn a commission at no extra cost to you. I only point to things I would use myself. I’m not a financial advisor, and nothing here is financial advice.

Two kinds of money in the same business

Most businesses have one pile of startup money: the stuff you buy. An ATM business has two.

The first pile is the machine. Retail-grade ATMs run a couple thousand dollars each, give or take, plus install and signage. That money is gone the day you spend it, and it comes back slowly through surcharges.

The second pile is the float: the cash you load into the machine so it has something to hand out. When a customer takes $60, the processor collects it from their bank and sends it back to you, usually the next banking day, with the fee on top. So the float is never lost. But it is never free either. It rotates, and while it rotates it is doing nothing else.

With $10,000, a sensible split looks roughly like this:

Bucket	Rough amount	What it does
Two machines, installed	~$5,000	Earns surcharges
Cash float in the cassettes	~$3,000–$3,500	Makes the machines possible; earns nothing itself
Reserve for repairs, slow months, weekend lag	~$1,500	Keeps you from panicking

That is two machines. Not five. Every guru video that says “start an ATM route with $10k” and shows a map full of pins has quietly skipped the second pile.

What the parked money costs you

I care about this more than most people would, because the whole philosophy of this blog is that money should be working for you. So let’s price the float honestly.

Say $3,500 sits in cassettes for a decade. If that same $3,500 had gone into a broad index fund and earned something like 7% a year (a historical-ish assumption, not a promise), it would grow to about $6,900 in ten years. That is roughly 1.07 to the tenth power, which is about 1.97, times $3,500. So the “free” float has a real opportunity cost of around $3,400 over ten years, before we even talk about the machines.

Even the boring option has a cost. At around 4% in a high-yield account, $3,500 earns about $140 a year. The float earns zero.

Does that make ATMs a bad idea? No. It means the machines have to clear a higher bar than their sticker price suggests. I wrote about this kind of math in compound interest is boring until you actually run the numbers, and it applies here in reverse: every dollar parked is a dollar not compounding.

The bar two machines have to clear

Here is the decision rule I would use. Two ATMs are worth it if, after 90 days of real data, they net meaningfully more than:

1. What the whole $10,000 would earn sitting in Treasury bills or a savings account (roughly $35 a month at 2026 rates), plus
2. A fair wage for your hours. Loading, logging, driving, fixing a jammed dispenser at 11pm. Call it a few hours per machine per month.

Most independently placed ATMs net somewhere around $100 to $900 a month each, and almost all of that spread comes from the location. A quiet laundromat might do a couple of withdrawals a day. A cash-only bar on a Friday might do more than that in an hour.

So a realistic two-machine start lands somewhere between “barely worth the gas” and “a few hundred dollars a month for a few hours of work.” The top of the range exists. It is not where most people start.

An illustrative path

To make this concrete, here is an illustrative example. Rob is a made-up high school teacher with summers off and $10,000 in savings. His numbers stay inside the ranges above.

Rob spends June walking into small businesses near his house: bars, a laundromat, two liquor stores, a bowling alley. He hears no a lot. By July he has two signed agreements, a liquor store and a bowling alley with a dead ATM from someone who gave up.

By October the two machines net a little over $400 a month combined. He loads the bowling alley on his way home Friday and the liquor store Monday morning. He keeps the reserve untouched in a savings account.

Here is the part I like. Rob does not buy a third machine in October. He waits until the reserve plus a few months of earnings covers it, so the third machine is funded by the first two instead of by more savings. That is the only way a small cash business grows without the owner feeling squeezed: the assets pay for the next asset.

And here is the part I would warn Rob about. He did not buy passive income. He bought a small job with good hours. That is fine, it might even be great, but it should be a clear-eyed choice.

Where I would be careful

Do not borrow the float

If you want to keep more of your savings liquid, financing the machines themselves on a 0% intro-APR business card can make sense, but only with a written plan to pay it off before the rate expires. Services like 7 Figures Credit help people line up that kind of business credit. The float is different. Funding cash that sits in a box with a cash advance is paying interest on money that is earning nothing. I would never do it.

The trend line is down

People use a little less cash every year. A good location today is probably a slightly less good location in five years. That is not a reason to skip ATMs, but it is a reason to want your payback period short and to treat each machine as something you might move.

Two machines is a test, not a business

The real value of the first $10,000 is information. After 90 days you know your market’s actual withdrawal numbers, what owners say yes to, and whether you like this kind of work. That is worth a lot. It is also not the same as financial freedom, and I would not tell my family it is.

ATMs versus the other things $10,000 can do

When I think about where to put a chunk like this, I line it up against the alternatives, the same way I did in vending machines vs real estate.

Option	Rough monthly result on $10k	Your time	The catch
Savings / T-bills	~$33–$37	None	Rates can drop
Index fund	~$0 paid out, grows over time	None	Can fall hard for years
Two vending machines	A few hundred, location-dependent	A few hours a week	Restocking, spoilage, location risk
Two ATMs	A few hundred, location-dependent	A few hours a month per machine	Float tied up, cash use declining

Vending and ATMs land in the same neighborhood. The honest differences are that vending ties no money up in float but asks more of your time (product, expiration dates, restocking), while ATMs need less hands-on time per machine but park cash in every box. If you hate inventory, ATMs are cleaner. If you hate idle capital, vending is.

I sold my vending business for reasons I wrote about in why I sold my vending machine business, and a lot of that reasoning applies to any route business: it is only as good as your locations, and it asks for more of your weekends than the videos admit.

I kept this post at the level of the decision. The nuts-and-bolts version lives on VendBuddy’s (full disclosure: VendBuddy is my company) blog: Start an ATM Business With $10k: Machines, Float and First 90 Days.

If you decide to do it

Here is the order I would follow:

1. Set up the LLC, EIN and a separate business bank account first. Keep the ATM money out of your personal accounts from day one.
2. Choose a processor before you choose machines, since it decides what hardware works and how fast settlement lands.
3. Build a list of 40 to 60 cash-heavy businesses close to home and go talk to owners in person.
4. Sign the location, then order the machine. Never the other way around.
5. Track withdrawals weekly for 90 days before you buy machine three.

For step three, VendBuddy (vendbuddy.io/app) will pull the bars, laundromats and convenience stores in your ZIP along with the owner’s contact info, and you can buy a small credit pack for one area instead of a monthly plan. If you want the full operator’s budget with a week-by-week plan, the team wrote it up in more detail on the VendBuddy blog.

And if you are not sure yet whether you want a business at all, or just want your money working, start with creating your financial freedom runway. Knowing how many months you can go without a paycheck changes how much risk any of this feels like.

The question I would sit with before signing anything: am I buying income, or am I buying a job I would actually enjoy? Both can be good answers. Just know which one it is.