A machine with many parents
Ask who invented the ATM and you will get several confident answers. A British engineer is often named. So is an American one. A Scottish engineer has a strong claim to a different part of the idea. Swedish banks also point to a machine that opened in the same summer as the British one. This is a good example of how “first” depends on definitions: first to be installed, first to work with a card, first to be connected to a bank’s computer, or first to be widely used?
An early experiment that did not catch on
Before the late 1960s, at least one machine tried to bring banking to a wall. Luther Simjian, an inventor known for many unrelated devices, developed the Bankograph, which was installed at a New York bank in 1960. It was reportedly meant to accept deposits of checks and cash and to issue a receipt. According to accounts, customers did not take to it and it was withdrawn within months. It is usually described as a forerunner rather than an ancestor of the modern machine, since it did not dispense cash.
The British machine, 1967
The date most often given for the first cash dispenser in service is June 27, 1967, at a Barclays branch in Enfield, north London. The machine was made by De La Rue and is often credited to John Shepherd-Barron, who worked for the firm. The widely repeated story is that the idea came to him while he was in the bath, thinking about how chocolate bars could be bought from a vending machine and cash could be dispensed in the same way.
How it worked differs from a modern ATM. Customers used paper vouchers or checks that were treated with a faintly radioactive material, reported as carbon-14. The machine matched the voucher against a personal identification number typed in on a keypad. Reports say the number was originally intended to have six digits, and was shortened to four after a suggestion from Shepherd-Barron’s wife, who said four would be easier to remember. This detail comes from his own later recollection, so treat it as a good story rather than a settled fact.
The machine was not tied to a bank’s central computer; it dispensed a fixed amount and gathered the vouchers inside. It was effectively a very secure vending machine for money.
The Scottish contribution
James Goodfellow, an engineer working for Smiths Industries in Scotland, developed a system that used a plastic card and a PIN, and received a British patent for it in 1966, according to usual accounts. His work is often described as the origin of the card-and-PIN approach. This is the part of the ATM that later became standard, so many writers present Goodfellow as the inventor of the PIN-based cash machine and Shepherd-Barron as the person whose machine opened first. Both statements can be true at once.
Sweden and the United States
Swedish savings banks introduced a cash dispenser, sometimes called the Bankomat, in Uppsala in mid-1967. It was developed by a Swedish firm, and it too used a form of card or token. Some sources treat its opening as nearly simultaneous with Enfield, so it is hard to give priority to one.
In the United States, a machine developed by Docutel, a Dallas company, is closely linked with Don Wetzel, who is often named as a principal inventor. The first Docutel machine to be installed is usually described as being at Chemical Bank in Rockville Center, New York, in 1969. This was a machine using a plastic card with a coded strip, and it had a degree of what we would now call an ATM’s functionality: it was designed to be used by the public at any time, with customers entering a code, selecting an amount and receiving cash. It is often presented as a template for the machines that followed in the United States.
Why the credit is shared
Several reasons explain why historians and companies rarely agree:
- Different features. One machine was first to be installed, another first to use a plastic card, another first to rely on a PIN with that card, and another first to be linked directly to account records.
- Parallel work. Engineers in different countries were drawn to the same problem at the same time, because banks wanted to reduce queueing and branch staff costs.
- Later storytelling. Companies and inventors told their own stories years afterward, and details of dates and motives were sometimes smoothed over.
- Patents versus prototypes. A filing date and an opening date are different events, and they sometimes point to different people.
A fair summary is that 1967 to 1969 was the period in which the first public machines appeared in Britain, Sweden and the United States, and that Shepherd-Barron, Goodfellow and Wetzel are each credited with something real.
What came next
These first machines were limited. Many could dispense only fixed amounts, some kept the card, and few communicated with a central system in real time. Over the 1970s, banks connected machines to their computers over telephone lines, which let customers check balances and make deposits, and which is the part of the story where wide area networks became important to everyday banking. The wider spread of the ATM, and the way card and PIN security evolved, is its own subject.
Why the story is called forgotten
The machines themselves were soon replaced. The vouchers, the carbon-14 and the fixed-amount dispensers vanished, and the modern ATM that people use is a descendant of several lines of work rather than of one. Remembering the details means resisting the neat single-inventor narrative that advertising and obituaries tend to prefer.