Cash or Code:
The Tech That Powers Your Grocery Run

One of my favorite innovations of the past few decades is the self-checkout at grocery stores. I love being able to scan my items and arrange them in my bags the way I prefer. Because getting groceries is such a mundane, weekly task, it's easy to go through the self-checkout line without thinking about the tools that actually make this whole process work. It turns out there’s a lot of science and technology powering the checkout experience.

If you ask your grandparents what it was like to get groceries when they were kids, their experience will sound a lot different than ours today. Before the 1960s, grocery workers had to price, inventory, and ring up all items in a store manually. Even cash registers were not completely electronic until the early 1970s. The checkout process took a long time, and both customers and workers were searching for better answers. In 1966, the Kroger Company (yes, that Kroger) sent out a plea for new technology to help their checkout and inventory systems. Their request was answered by the Radio Corporation of America (RCA). While searching for ways to make checkout faster, researchers at RCA came across a patent for the barcode submitted by Joe Woodland and Bernard Silver in 1949. They paired this idea with laser technology developed in the 1960s by Theodore Maiman to create an automated checkout system for grocery stores. The first trial run of these systems was conducted in 1972, and it was a roaring success. Around this time, IBM also hopped on the checkout conveyor belt to help develop a universal barcode system that could be used in any grocery store. Through their efforts, the Universal Product Code (UPC) was invented and is still in use today. The first UPC was scanned (on a pack of Wrigley’s chewing gum) in 1974, and by 1989, over half of all grocery sales used UPC scanning. 

So how exactly do these barcodes and scanners work? Barcodes are typically composed of black bars against a white background. Similar to how Morse code uses dashes and dots of different lengths to convey different letters, the width of each line within a barcode corresponds to a different number. In fact, the inventor of the barcode, Joe Woodland, was actually inspired by Morse code when he came up with the idea. While we can train to learn Morse code, it can be difficult to tell differences in line widths on a barcode with the human eye. Thus, we need a scanner to “read” the barcode for us. Scanners have two main components: a light source that emits a laser onto the barcode and a light sensor that detects the light that bounces back. When the light from the laser hits the black and white lines on a barcode, the white portion reflects the light back and the black portion absorbs the light. This creates a pattern of light reflected back to the sensor on the scanner. This pattern is then converted into a series of numbers and transmitted to the computer system that the scanner is hooked up to. A scanner’s accuracy depends on its light being bright and focused enough to get a reliable signal to bounce back from the barcode. This light requirement is why the barcode could not be used until the laser was invented. Newer versions of barcodes, like QR codes and other 2D barcodes, rely on cameras to capture the image, enabling these codes to be more complicated and to be detected without the need of a laser beam.

Now that you’ve scanned all the items in your cart, it’s time to pay. Like most people in 2025, you’re probably not using cash, so you need to use the payment terminal. Similar to barcode scanners, the first card reader came out in the 1960s. Credit cards were introduced decades before, but their usage relied on manual information input and payment approval via telephone. Once again, IBM came to the rescue with new-and-improved cards and machines to use them. This time they weren’t using lasers to read information–they employed magnets. The use of magnetized material to store information started during WWII, but by the 1950s, computing companies like IBM started using magnetics for disk storage. They transferred this idea to credit cards, where they added a magnetic strip onto the back to hold information. How does a magnet hold information, you might ask? Just like the barcode, it all comes down to patterns. The strip on the back of a credit card contains iron. This iron is initially demagnetized, meaning it has no magnetic field and acts as  a blank canvas for information to be imprinted upon. An electrical current is applied to the strip in a pattern that corresponds to a series of numbers (e.g., the credit card number). Thus, the iron particles on the strip are magnetized in this specific pattern that can be read by a machine. Within a card reader, there is a “read head” that detects magnetic fields. When you swipe the magnetic strip across it, the read head senses the pattern of magnetization along the strip. This pattern is then converted into numbers by other electronics within the reader, and your credit card information can be processed by the register.

It’s likely been a while since you’ve actually swiped your credit or debit card at a payment terminal. That’s because safety concerns about magnetic-stripe cards were raised in the 80s and 90s. The magnetic signals on these cards could be “skimmed” by scammers who attached small probes to card readers, making it easy for credit card information to be stolen and replicated. Luckily, a new technology emerged in the mid-1980s to improve card safety: chip cards. Initially called smart-cards, these credit and debit cards are embedded with an integrated circuit chip (also called EMV chips). These chips are miniprocessors that include secure memory storage for card information, a cryptographic engine that makes unique codes (called cryptograms), and a tiny operating system to run instructions during each transaction. Yes, you read that right - there’s a mini computer inside of your credit card! When you insert the chip into the payment terminal, metal contacts inside the reader power up the small computer chip. The chip’s cryptographic engine then creates a cryptogram for that specific transaction. This code is sent to your bank, which decrypts the information and either approves or denies the transaction. This process is much safer than magnetic storage because the chip creates a unique cryptogram for each transaction and your card information is never sent directly to the payment terminal. Even if a hacker were to intercept a cryptogram for a payment, it would be useless in the future because each transaction gets its own code. As a backup, cards often still have magnetic strips, but using the chip is the recommended way to pay when using card readers. 

While the chip makes transactions safer, it typically takes more time than swiping the magnetic stripe. If we learned anything from our predecessors in the 1960s, we want this process to be fast and efficient. Enter in, the “tap to pay” option. In addition to being tiny computers inside your credit card, chips are enabled for radio frequency identification (RFID). This allows the chip to send information through radio waves instead of through electrical contact. When the chip comes into the range of the terminal, the card reader emits electromagnetic waves to temporarily power up the microprocessor (the same way a wireless phone charger works). The chip then does its work to create a one-time code for the transaction and sends the information through a wire contained in the credit card. This wire serves as an antenna, emitting radio waves to the card reader, which picks up the signal and receives the payment information. 

If you forgot your credit card at home, you can also use your phone as payment. Similar to tapping your credit card, smartphones are able to use a type of RFID called near field communication (NFC). The principle is the same, but instead of a small microprocessor sending information via radiowaves, it’s the processors already inside of your smartphone. You simply have to add your card information into a secure app on your phone, and it will be able to create cryptograms and interact with a card reader just like a chip card.  

Grocery store checkouts have greatly evolved over time, and they will continue to do so as technology improves. It may even be possible to cut out entire steps of the checkout process, such as scanning items for purchase. This is the case at Amazon Go stores, where items are tracked as customers take them off of shelves. At the end of the trip, patrons simply walk out of the store, and Amazon charges them for their items using a saved payment method. When I first heard about these stores, I was amazed but also skeptical. These stores sound super convenient, but I had a litany of questions as well: how are customers being tracked? Is the tracking information stored somehow, and if so, what is Amazon doing with that information? How accurate is this tracking? How safe is it? As I asked these questions, it struck me that people in the 1960s probably felt the same way about scanners which are a mundane part of life now. Given the rapid pace of technology evolution in our day and age, we may not have as much time to adjust to changes in our checkout routine, but it will happen regardless. For now, I will continue to use the standard self-checkout with a new appreciation for all the technology and innovation that went into making it an efficient (and enjoyable) process for me.