A shopper puts a bottle of shampoo on the belt, the cashier passes it over the scanner, there is a beep, and the name and price appear on the screen. It feels instant and simple. Behind that beep sit a few distinct steps, and understanding them answers the questions brands now ask most often: will a 2D code work at the till, does the checkout go online, and where does the price come from?
Data carriers compared
| Carrier | What it holds | Who reads it | Typical use | Opens on a phone? |
|---|---|---|---|---|
| Linear barcode (EAN-13 / UPC-A) | GTIN only | Laser and imaging scanners at checkout | Retail checkout today; nearly universal | Not useful to consumers |
| Plain QR code | Any text, usually a web address | Phones; not interpreted as a product by POS | Marketing links, campaigns | Yes — opens a web page |
| QR code with GS1 Digital Link | A web address that contains the GTIN and optional batch, serial, expiry | Phones, and 2D-capable POS scanners that extract the GTIN | Consumer-facing packs that also need to scan at checkout | Yes — resolves to brand content |
| GS1 DataMatrix | GS1 element string: GTIN plus batch, serial, expiry and more | 2D imaging scanners; most phone camera apps will not open it | Healthcare, regulated and small packs, industrial marking | Usually not, without a dedicated app |
The short version: four steps
Every scan at a point of sale (POS) follows the same basic sequence:
| Step | What happens | Where it happens |
|---|---|---|
| 1. Read | The scanner captures the barcode | The scanner |
| 2. Decode | The symbol is turned into data, and the product's GTIN is extracted | The scanner or its software |
| 3. Look up | The till finds that GTIN in the retailer's item file | The retailer's own systems |
| 4. Record | The sale is added to the basket and, later, to stock and sales records | The retailer's own systems |
The important point is in the last column. Steps 3 and 4 use data the retailer already holds. The barcode supplies an identifier. It does not supply the product name or the price.
Step 1: the scanner reads the code
There are two broad families of checkout scanner.
Laser scanners sweep a beam of light across a barcode and measure the reflections from its bars and spaces. They work well with traditional linear (1D) barcodes such as EAN/UPC, the familiar stripes on nearly every retail pack. They cannot read 2D barcodes at all.
Imaging scanners (imagers) take a picture of the code, much like a camera, and decode it in software. They can read linear barcodes and 2D barcodes, the square codes that store data in a grid of dots or modules.
This distinction drives the retail industry's move to 2D. The industry goal known as Ambition 2027 (often called Sunrise 2027) is for checkouts worldwide to be able to read and process the product number from both linear and 2D barcodes by the end of 2027. GS1 notes that meeting it requires software upgrades, and may require new equipment where scanners are not imager-capable. It is an industry objective, not a law. See what Sunrise 2027 really means.
Step 2: the scanner turns the symbol into data
What the scanner gets out of the code depends on what the code carries.
A linear EAN/UPC barcode carries one thing: the GTIN (Global Trade Item Number), the number that identifies the product type. Every bottle of the same shampoo in the same size carries the same GTIN.
A GS1 DataMatrix stores data as GS1 element strings. Each piece of data is introduced by a short numeric Application Identifier (AI) that says what it is: (01) for the GTIN, (10) for the batch or lot, (17) for the expiry date, (21) for a serial number. The scanner can pass the GTIN alone, or the GTIN plus the other fields, depending on how it is set up.
A QR code carrying a GS1 Digital Link stores a web address that has the GS1 identifiers built into it, for example:
https://example.com/01/09506000134376/10/AB123?17=271231
To the checkout, this is not a link to follow. It is structured data to unpack. The scanner or its software converts the address into GS1 element strings and passes the GTIN to the till. GS1 states that Digital Link identifiers can be "decoded by scanning software and passed to the host system without any need for an online connection." For this to work at retail, the Digital Link in the code must be in its uncompressed form.
Those three carriers — a GS1 DataMatrix, a QR code with a GS1 Digital Link URI, and a Data Matrix with a GS1 Digital Link URI — are the 2D options GS1 approves for general retail checkout. A "GS1 QR Code" using element-string syntax is not one of them. The differences are compared in QR code vs GS1 DataMatrix.
What about packs with two codes?
During the transition, GS1 rules require the linear barcode to stay on general retail packs alongside any 2D code. Both must encode the same GTIN, and any shared values must match. The checkout is expected to give one beep and process one set of data per item, so a pack with two codes is not rung up twice.
Scanners can be configured to handle this in different ways:
- Mode 1, the minimum for Ambition 2027: send the first GTIN read from either barcode.
- Mode 2: prefer the 2D barcode and fall back to the linear one.
- Mode 3: read all compliant barcodes.
Which mode a retailer uses is its own decision. Why both codes stay for now is explained in why 1D and 2D barcodes will coexist.
Step 3: the till looks up the GTIN
Once the till has the GTIN, it searches the retailer's own item file: a product database keyed by GTIN, which the retailer built when it agreed to stock the product. That record holds the short description printed on the receipt, the selling price for that store, any active promotion, tax treatment and often the retailer's own internal item code.
The till does not ask GS1 for anything. It does not contact the brand. It does not open the web address in a Digital Link QR code. GS1 puts it plainly: retailers use the barcode "to look up their own information in their IT systems, exactly as they do today."
That is why the same product can have different prices in two shops on the same street: each retailer holds its own record. Where those records come from is covered in where retailers get the product name and price, and the brand-side data flow in how product information gets from a brand to a retailer.
Step 4: the sale is recorded
When the basket is paid for, the retailer's systems record the sale against the GTIN. That data feeds stock levels, reordering, sales reports and, in many retailers, loyalty schemes. All of it lives in the retailer's systems. None of it is sent to GS1 as part of the scan.
What changes when the code is 2D
For pricing, nothing changes: the till still looks up the GTIN in the same item file. What changes is what else the retailer could read.
A linear EAN/UPC barcode can only say "this is a 200 ml bottle of shampoo". A 2D code can also say "from batch AB123, expiry date 31 December 2027". A retailer whose scanners and software capture those extra fields could, if it chooses, use them for things such as:
- Blocking expired items at the till, by comparing the expiry date with today's date and stopping the sale.
- Blocking a recalled batch, so that only the affected lot is stopped rather than every unit of the product.
- Automatic markdowns for items close to their expiry date.
- Better stock rotation and waste data, because sales are recorded by batch as well as by product.
These are possibilities, not universal practice. The industry minimum is reading the GTIN; capturing batch or expiry depends on each retailer's equipment, software and processes. We look at each use in how retailers can use GTIN, batch, serial and expiry.
A worked example
Lumen Dairy, a fictional brand, puts a QR code carrying a GS1 Digital Link on its yoghurt pots, next to the existing EAN-13 barcode. The QR code holds the GTIN, the batch and the expiry date.
- At Northgate Market, a supermarket with older laser scanners, the scanner reads the EAN-13. The till gets the GTIN, finds Northgate's record and charges Northgate's price. The QR code is simply ignored.
- At Corvina Foods, a chain that has upgraded to imaging scanners in Mode 2, the scanner prefers the QR code. Its software unpacks the Digital Link, passes the GTIN to the till, and the till charges Corvina's price from Corvina's item file. One beep, one item.
- Corvina has also chosen to capture expiry. When a pot past its date reaches the till, the system flags it and the cashier removes it from the sale.
- At home, the shopper points a phone camera at the same QR code. The phone opens the web address, Lumen Dairy's resolver sends it to a recipe page, and Lumen Dairy's platform records the scan. This path, which does use the internet, is described in what happens when a consumer scans a connected product.
Neither checkout went online to read the code. Both priced the yoghurt from their own data.
What this means for brands
- Keep the linear barcode. GS1 rules require it alongside the 2D code during the transition, placed close to it.
- Encode the same GTIN in both codes, and make sure batch or expiry values in the 2D code match anything printed elsewhere on the pack.
- Print to grade, not just to scan. Retail barcodes, linear and 2D, have a minimum print quality grade of 1.5 (C) under ISO grading. A code that scans once on a phone may still fail at a busy checkout. See why printed is not the same as verified.
- Use an approved carrier, uncompressed. For a code that also serves consumers, that means a QR code (or Data Matrix) with an uncompressed GS1 Digital Link URI. How the same code serves both audiences is explained in why the same GS1 2D code can work for checkout and consumers.
- Get your product into each retailer's item file before it ships. No 2D code fixes a missing item record.